Curved sensor formed from silicon fibers
Methods and apparatus for fabricating a curved sensor from silicon fibers are disclosed. In another embodiment, a curved sensor is produced having mini-sensors of differing sizes. In another embodiment, these mini-sensors are configured so that every other mini-sensor in a row of mini-sensors is shifted upwards slightly relative to its neighboring mini-sensors in the same row.
The Present Continuation-in-Part Patent Application is related to:
-
- Pending U.S. Non-Provisional application Ser. No. 12/930,165, filed on 28 Dec. 2010;
- Pending U.S. Non-Provisional application Ser. No. 12/655,819, filed on 6 Jan. 2010;
- Provisional Patent Application 61/208,456, filed on 23 Feb. 2009, now abandoned; and
- Pending PCT International Patent Application No. PCT/GB2010/000298, filed on 19 Feb. 2010.
In accordance with the provisions of Sections 119 and/or 120 of Title 35 of the United States Code of Laws, the Applicants claim the benefit of priority for any and all subject matter which is commonly disclosed in the Present Continuation-in-Part Patent Application, and in any of the related Patent Applications identified above.
The subject matter of the Non-Provisional Applications and the PCT International Patent Applications identified above are hereby incorporated by reference.
FIELD OF THE INVENTIONOne embodiment of the present invention relates to a curved concave sensor formed from silicon fibers.
FEDERALLY SPONSORED RESEARCH OR DEVELOPMENTNone.
BACKGROUND OF THE INVENTION I. A Brief History of Cameras Evolution of the Three Primary Camera TypesCurrent photographic cameras evolved from the first “box” and “bellows” models into three basic formats by the late twentieth century.
The rangefinder came first. It was followed by the SLR, or, single lens reflex and finally the Compact “Point and Shoot” cameras. Most portable cameras today use rangefinder, SLR or “Point and Shoot” formats.
Simple Conventional CamerasA simple lens with a flat film or sensor faces several problems. Light travels over a longer pathway to the edges of the film or the sensor's image area, diluting those rays. Besides being weaker, as those rays travel farther to the sensor's edges, they suffer more “rainbow effect,” or chromatic aberration.
The eagle's retina has eight times as many retinal sensors as the human eye. They are arranged on a sphere the size of a marble. The eagle's rounded sensors make simpler optics possible. No commercially available camera that is available today has a pixel count which equals a fourth of the count of sensors in an eagle's eye. The eagle eye uses a simple lens and a curved retina. The best conventional cameras use multiple element lenses with sophisticated coatings, rare earth materials and complex formulas. This is all to compensate for their flat sensors. The eagle sees clearly at noon, in daylight or at dusk with simpler, lighter and smaller optics than any camera.
Rangefinder CamerasRangefinder cameras are typified by a broad spectrum from the early LEICA™ thirty-five millimeter cameras, for professionals, to the later “INSTAMATIC™” film types for the masses. (Most of KODAK' s™ INSTAMATIC™ cameras did not focus, so they were not true rangefinders. A few “Instamatic type” models focused, and had a “viewing” lens separated from the “taking” lens, qualifying them as rangefinders.)
Rangefinder cameras have a “taking” lens to put the image on the film (or sensor today) when the shutter opens and closes; mechanically or digitally. These cameras use a second lens for viewing the scene. Focusing takes place through this viewing lens which connects to, and focuses, the taking lens.
Since the taking lens and the viewing lens are different, and have different perspectives on the scene being photographed, the taken image is always slightly different than the viewed image. This problem, called parallax, is minor in most situations but becomes acute at close distances.
Longer telephoto lenses, which magnify more, are impractical for rangefinder formats. This is because two lenses are required, they are expensive and require more side-to-side space than exists within the camera body. That's why no long telephoto lenses exist for rangefinder cameras.
Some rangefinder cameras use a frame in the viewfinder which shifts the border to match that of the taking lens as the focus changes. This aligns the view with the picture actually taken, but only for that portion that's in focus. Backgrounds and foregrounds that are not in focus shift, so those parts of the photographed image still vary slightly from what was seen in the viewfinder.
A few rangefinder cameras do exist that use interchangeable or attachable lenses, but parallax remains an unsolvable problem and so no manufacturer has ever successfully marketed a rangefinder camera with much beyond slightly wide or mildly long telephoto accessories. Any added rangefinder lens must also be accompanied by a similar viewfinder lens. If not, what is viewed won't match the photograph taken at all. This doubles the lens cost.
A derivation of the rangefinder, with the same limitations for accessory lenses, was the twin lens reflex, such as those made by ROLLEI-WERKE™ cameras.
Compact, or “Point and Shoot” CamerasCurrently, the most popular format for casual photographers is the “Point and Shoot” camera. They emerged first as film cameras but are now nearly all digital. Many have optical zoom lenses permanently attached with no possibility for interchanging optics. The optical zoom, typically, has a four to one range, going from slight wide angle to mild telephoto perspectives. Optical zooms don't often go much beyond this range for acceptable results and speed. Some makers push optical zoom beyond this four to one range, but the resulting images and speeds deteriorate. Others add digital zoom to enhance their optical range; causing results that most trade editors and photographers currently hate, for reasons described in following pages.
There are no “Point and Shoot” cameras with wide angle lenses as wide as the perspective are for an eighteen millimeter SLR lens (when used, for relative comparison, on the old standard thirty-five millimeter film SLR cameras.) There are no “Point and Shoot” cameras with telephoto lenses as long as a two hundred millimeter SLR lens would have been (if on the same old thirty-five millimeter film camera format.)
Today, more photographs are taken daily by mobile phones and PDAs than by conventional cameras. These will be included in the references herein as “Point and Shoot Cameras.”
Single Lens Reflex (SLR) CamerasSingle lens reflex cameras are most commonly used by serious amateurs and professionals today since they can use wide selections of accessory lenses.
With 35 mm film SLRs, these lenses range from 18 mm “fisheye” lenses to 1,000 mm super-telephoto lenses, plus optical zooms that cover many ranges in between.
With SLRs there's a mirror behind the taking lens which reflects the image into a viewfinder. When the shutter is pressed, this mirror flips up and out of the way, so the image then goes directly onto the film or sensor. In this way, the viewfinder shows the photographer almost the exact image that will be taken, from extremes in wide vistas to distant telephoto shots. The only exception to an “exact” image capture comes in fast action photography, when the delay caused by the mirror movement can result in the picture taken being slightly different than that image the photographer saw a fraction of a second earlier.
This ability to work with a large variety of lenses made the SLR a popular camera format of the late twentieth century, despite some inherent disadvantages.
Those SLR disadvantages are the complexity of the mechanism, requiring more moving parts than with other formats, plus the noise, vibration and delay caused by the mirror motion. Also, lens designs are constrained, due to the lens needing to be placed farther out in front of the path of the moving mirror, which is more distant from the film or sensor, causing lenses to be heavier, larger and less optimal. There is also the introduction of dust, humidity and other foreign objects into the camera body and on the rear lens elements when lenses are changed.
Dust became a worse problem when digital SLRs arrived, since the sensor is fixed, unlike film. Film could roll away the dust speck so only one frame was affected. With digital cameras, every picture is spotted until the sensor is cleaned. Recent designs use intermittent vibrations to clear the sensor. This doesn't remove the dust from the camera and fails to remove oily particles. Even more recent designs, recognizing the seriousness of this problem, have adhesive strips inside the cameras to capture the dust if it is vibrated off the sensor. These adhesive strips, however, should be changed regularly to be effective, and, camera users typically would require service technicians to do this.
Some of these “vibration” designs assume all photos use a horizontal format, with no adhesive to catch the dust if the sensor vibrates while in a vertical position, or, when pointed skyward or down.
Since the inherent function of an SLR is to use interchangeable lenses, the problem continues.
Extra weight and bulk are added by the mirror mechanism and viewfinder optics to SLRs. SLRs need precise lens and body mounting mechanisms, which also have mechanical and often electrical connections between the SLR lens and the SLR body. This further adds weight, complexity and cost.
Optical Zoom LensesOptical zoom lenses reduce the need to change lenses with an SLR. The photographer simply zooms in or out for most shots. Still, for some situations, an even wider or longer accessory lens is required with the SLR, and the photographer changes lenses anyway.
Many “Point and Shoot” cameras today have zoom lenses as standard; permanently attached. Nearly all SLRs offer zoom lenses as accessories. While optical technology continues to improve, there are challenges to the zoom range any lens can adequately perform. Other dilemmas with zoom lenses are that they are heavier than their standard counterparts, they are “slower,” meaning less light gets through, limiting usefulness, and zoom lenses never deliver images that are as sharp or deliver the color fidelity as a comparable fixed focal length lens. And again, the optical zoom, by moving more elements in the lens, introduces more moving parts, which can lead to mechanical problems with time and usage, plus added cost. Because optical zooms expand mechanically, they also function like an air pump, sucking in outside air while zooming to telephoto and squeezing out air when retracting for wider angle perspectives. This can introduce humidity and dust to the inner elements.
II. The Limitations of Conventional Mobile Phone CamerasThe Gartner Group reported that over one billion mobile phones were sold worldwide in 2009. A large portion of currently available mobile phones include a camera. These cameras are usually low quality photographic devices with simple planar arrays situated behind a conventional lens. The quality of images that may be captured with these cell phone cameras is generally lower than that which may be captured with dedicated point-and-shoot or more advanced cameras. Cell phone cameras usually lack advanced controls for shutter speed, telephoto or other features.
Conventional cell phone and PDA cameras suffer from the same four deficiencies.
-
- 1. Because they use flat digital sensors, the optics are deficient, producing poor quality pictures. To get normal resolution would require larger and bulkier lenses, which would cause these compact devices to become unwieldy.
- 2. Another compromise is that these lenses are slow, gathering less light. Many of the pictures taken with these devices are after sunset or indoors. This often means flash is required to enhance the illumination. With the lens so close to the flash unit, as is required in a compact device, a phenomena known as “red-eye” often occurs. (In darkened situations, the pupil dilates in order to see better. In that situation, the flash often reflects off the subject's retina, creating a disturbing “red eye” image. This is so common that some camera makers wired their devices so a series of flashes go off before the picture is taken with flash, in an attempt to close down the pupils. This sometimes works and always disturbs the candid pose. Pencils to mark out “red eye” are available at retail. There are “red eye” pencils for humans and even “pet eye” pencils for animals. Some camera software developers have written algorithms that detect “red eye” results and artificially remove the “red eye,” sometimes matching the subject's true eye color, but not always.
- 3. Flash photography shortens battery life.
- 4. Flash photography is artificial. Faces in the foreground can be bleached white while backgrounds go dark. Chin lines are pronounced, and it sometimes becomes possible to see into a human subject's nostrils, which is not always pleasing to viewers.
Current sales of high definition television sets demonstrate the growing public demand for sharper images. In the past, INSTAMATIC® cameras encouraged more picture-taking, but those new photographers soon tired of the relatively poor image quality. Thirty-five millimeter cameras, which were previously owned mostly by professionals and serious hobbyists, soon became a mass market product.
With unprecedented numbers of photos now being taken with mobile phones, and the image quality being acceptable now, but probably less acceptable in the near future, this cycle is likely to repeat.
The development of a system that reduces these problems would constitute a major technological advance, and would satisfy long-felt needs in the imaging business.
SUMMARY OF THE INVENTIONThe present invention comprises methods and apparatus for a curved sensor which is manufactured from silicon or other fibers.
In another embodiment of the invention, pixels having varying sizes are formed on a sensor.
In yet another embodiment, every other pixel in each ring of pixels is positioned slightly above its two neighboring pixels.
An appreciation of the other aims and objectives of the present invention, and a more complete and comprehensive understanding of this invention, may be obtained by studying the following description of a preferred embodiment, and by referring to the accompanying drawings.
The present invention provides methods and apparatus related to a camera having a non-planar or curved sensor. The present invention may be incorporated in a mobile communication device. In this Specification, and in the Claims that follow, the terms “mobile communication device” and “mobile communication means” are intended to include any apparatus or combination of hardware and/or software which may be used to communicate, which includes transmitting and/or receiving information, data or content or any other form of signals or intelligence.
Specific examples of mobile communication devices include cellular or wireless telephones, smart phones, personal digital assistants, laptop or netbook computers, iPads™ or other readers/computers, or any other generally portable device which may be used for telecommunications or viewing or recording visual content.
Unlike conventional cellular telephones which include cameras that utilize conventional flat sensors, the present invention includes curved or otherwise non-planar sensors. In one embodiment, the non-planar surfaces of the sensor used in the present invention comprise a plurality of small flat segments which altogether approximate a curved surface. In general, the sensor used by the present invention occupies three dimensions of space, as opposed to conventional sensors, which are planes that are substantially and generally contained in two physical dimensions.
The present invention may utilize sensors which are configured in a variety of three-dimensional shapes, including, but not limited to, spherical, paraboloidal and ellipsoidal surfaces.
In the present Specification, the terms “curvilinear” and “curved” encompass any line, edge, boundary, segment, surface or feature that is not completely colinear with a straight line. The term “sensor” encompasses any detector, imaging device, measurement device, transducer, focal plane array, charge-coupled device (CCD), complementary metal-oxide semiconductor (CMOS) or photocell that responds to an incident photon of any wavelength.
While some embodiments of the present invention are configured to record images in the optical spectrum, other embodiments of the present invention may be used for a variety of tasks which pertain to gathering, sensing and/or recording other forms of radiation. Embodiments of the present invention include systems that gather and/or record color, black and white, infra-red, ultraviolet, x-rays or any other stream of radiation, emanation, wave or particle. Embodiments of the present invention also include systems that record still images or motion pictures.
II. Specific Embodiments of the InventionEmbodiments of the present invention include, but are not limited to, mobile communication devices with a camera that incorporate the following sensors:
-
- 1. Curved sensors: Generally continuous portions of spheres, or revolutions of conic sections such as parabolas or ellipses or other non-planar shapes. Examples of a generally curved sensor 12 appear in
FIGS. 4A , 4B and 4C. In this specification, various embodiments of curved sensors are identified with reference character 12, 12a, 12b, 12c, and so on. - 2. Faceted sensors: Aggregations of polygonal facets or segments. Any suitable polygon may be used, including squares, rectangles, triangles, trapezoids, pentagons, hexagons, septagons, octagons or others.
FIG. 5 exhibits a sensor 12a comprising nine flat polygonal segments or facets 32a. For some applications, a simplified assembly of a few flat sensors might lose most of the benefit of a smoother curve, while achieving a much lower cost.FIGS. 6 and 7 provide side and perspective views of a generally spherical sensor surface 12b comprising a number of flat facets 32b.FIG. 7 shows exaggerated gaps 34 between the facets. The facets could each have hundreds, thousands or many millions of pixels. In this specification, the facets of the sensor 12 are identified with reference characters 32, 32a, 32b, 32c and so on.
- 1. Curved sensors: Generally continuous portions of spheres, or revolutions of conic sections such as parabolas or ellipses or other non-planar shapes. Examples of a generally curved sensor 12 appear in
In one alternative embodiment, a flat center sensor might be surrounded by these “petals” with squared-off points.
Digital zoom is software-driven. The camera either captures only a small portion of the central image, the entire scene or any perspective in between. The monitor shows the operator what portion of the overall image is being recorded. When digitally zooming out to telephoto in one embodiment of the present invention, which uses denser pixels in its center, the software can use all the data. Since the center has more pixels per area, the telephoto image, even though it is cropped down to a small section of the sensor, produces a crisp image. This is because the pixels are more dense at the center.
When the camera has “zoomed back” into a wide angle perspective, the software can compress the data in the center to approximate the density of the pixels in the edges of the image. Because so many more pixels are involved in the center of this wide angle scene, this does not effect wide angle image quality. Yet, if uncompressed, the center pixels represent unnecessary and invisible detail captured, and require more storage capacity and processing time. Current photographic language might call the center section as being processed “RAW” or uncompressed when shooting telephoto but being processed as “JPEG” or other compression algorithm in the center when the image is wide angle.
Digital zoom is currently disdained by industry experts. When traditional sensors capture an image, digital zooming creates images that break up into jagged lines, forms visible pixels and yields poor resolution.
Optical zoom has never created images as sharp as fixed focus length lenses are capable of producing. Optical zooms are also slower, letting less light through the optical train.
Embodiments of the present invention provide lighter, faster, cheaper and more dependable cameras. In one embodiment, the present invention provides digital zoom. Since this does not require optical zoom, it uses inherently lighter lens designs with fewer elements.
In various embodiments of the invention, more pixels are concentrated in the center of the sensor, and fewer are placed at the edges of the sensor. Various densities may be arranged in between the center and the edges. This feature allows the user to zoom into a telephoto shot using the center section only, and still have high resolution.
In one embodiment, when viewing the photograph in the wide field of view, the center pixels are “binned” or summed together to normalize the resolution to the value of the outer pixel density.
When viewing the photograph in telephoto mode, the center pixels are utilized in their highest resolution, showing maximum detail without requiring any adjustment of lens or camera settings.
The digital zoom feature offers extra wide angle to extreme telephoto zoom. This feature is enabled due to the extra resolving power, contrast, speed and color resolution lenses are able to deliver when the digital sensor is not flat, but curved, somewhat like the retina of a human eye. The average human eye, with a cornea and single lens element, uses, on average, 25 million rods and 6 million cones to capture images. This is more image data than is captured by all but a rare and expensive model or two of the cameras that are commercially available today, and those cameras typically must use seven to twenty element lenses, since they are constrained by flat sensors. These cameras cannot capture twilight images without artificial lighting, or, by boosting the ISO which loses image detail. These high-end cameras currently use sensors with up to 48 millimeter diagonal areas, while the average human eyeball has a diameter of 25 millimeters. Eagle eyes, which are far smaller, have eight times as many sensors as a human eye, again showing the optical potential that a curved sensor or retina provides. Embodiments of the present invention are more dependable, cheaper and provide higher performance. Interchangeable lenses are no longer necessary, which eliminates the need for moving mirrors and connecting mechanisms. Further savings are realized due to simpler lens designs, with fewer elements, because flat film and sensors, unlike curved surfaces, are at varying distances and angles from the light coming from the lens. This causes chromatic aberrations and varying intensity across the sensor. To compensate for that, current lenses, over the last two centuries, have mitigated the problem almost entirely, but, with huge compromises. Those compromises include limits on speed, resolving power, contrast, and color resolution. Also, the conventional lens designs require multiple elements, some aspheric lenses, exotic materials and special coatings for each surface. Moreover, there are more air to glass surfaces and more glass to air surfaces, each causing loss of light and reflections.
Variable Density of PixelsIn some embodiments of the present invention, the center of the sensor, where the digitally zoomed telephoto images are captured, is configured with dense pixilation, which enables higher quality digitally zoomed images.
In another embodiment of the invention, suitable software compresses the dense data coming from the center of the image when the camera senses that a wide angle picture is being taken. This feature greatly reduces the processing and storage requirements for the system.
Lens ShadeOther embodiments of the invention include a lens shade, which senses the image being captured, whether wide angle or telephoto. When the camera senses a wide angle image, it retracts the shade, so that the shade does not get into the image area. When it senses the image is telephoto, it extends, blocking extraneous light from the non-image areas, which can cause flare and fogged images.
Embodiments of the present invention reduce the dust problem that plagues conventional cameras since no optical zoom or lens changes are needed. Accordingly, the camera incorporated into the mobile communication device is sealed. No dust enters to interfere with image quality. An inert desiccated gas, such as Argon, Xenon or Krypton may be sealed in the lens and sensor chambers within the enclosure 14, reducing oxidation and condensation. If these gases are used, the camera also gains benefits from their thermal insulating capability and temperature changes will be moderated, and the camera can operate over a wider range of temperatures.
Completely Sealed CamerasIn another embodiment of the invention, the entire camera may be sealed with an inert gas, such as Argon, Krypton or Xenon.
Improved Optical PerformanceThe present invention may be used in conjunction with a radically high speed lens, useable for both surveillance without flash (or without floods for motion) or fast action photography. This becomes possible again due to the non-planar sensor, and makes faster ranges like a f/0.7 or f/0.35 lens designs, and others, within practical reach, since the restraints posed by a flat sensor (or film) are now gone.
All these enhancements become practical since new lens formulas become possible. Current lens design for flat film and sensors must compensate for the “rainbow effect” or chromatic aberrations at the sensor edges, where light travels farther and refracts more. Current lens and sensor designs, in combination with processing algorithms, have to compensate for the reduced light intensity at the edges. These compensations limit the performance possibilities.
Since the camera lens and body are sealed, an inert gas like Argon, Xenon or Krypton may be inserted, e.g., injected during final assembly, reducing corrosion and rust. The camera can then operate in a wider range of temperatures. This is both a terrestrial benefit, and, is a huge advantage for cameras installed on satellites.
Rotating & Shifted SensorsA motion camera can do the same, or, in a different embodiment, can simply move the sensor and capture only the new image using the data from the prior position to fill in the gaps in a continuous process.
This method captures an image using a moveable sensor with gaps between the sensors in its array of sensors. This method makes fabricating much easier, because the spaces between segments become less critical. So, in one example, a square sensor in the center is surrounded by a row of eight more square sensors, which, in turn, is surrounded by another row of sixteen square sensors. The sensors are sized to fit the circular optical image, and each row curves in slightly more, creating the non-planar total sensor.
In use, the camera first takes one picture. The sensor immediately rotates or shifts slightly and a second image is immediately captured. Software can tell where the gaps were and stitches the new data from the second shot into the first. Or, depending on the sensor's array pattern, it may shift linearly in two dimensions, and possibly move in an arc in the third dimension to match the curve.
This concept makes the production of complex sensors easier. The complex sensor, in this case, is a large sensor comprising multiple smaller sensors. When such a complex sensor is used to capture a focused image, the gaps between each sensor lose data that is essential to make the complete image. Small gaps reduce the severity of this problem, but smaller gaps make the assembly of the sensor more difficult. Larger gaps make assembly easier and more economical, but, create an even less complete image. The present method, however, solves that problem by moving the sensor after the first image, and taking a second image quickly. This gives the complete image and software can isolate the data that is collected by the second image that came from the gaps and splice it into the first image.
The same result may be achieved by a moving or tilting lens element or a reflector that shifts the image slightly during the two rapid sequence exposures. In this embodiment, the camera uses, but changes in a radical way, an industry technique known as “image stabilization.” The camera may use image stabilization in both the first and second images. This method neutralizes the effect of camera motion during an exposure. Such motion may come from hand tremors or engine vibrations. However, in this embodiment, after the first exposure, the camera will reverse image stabilization and introduce “image de-stabilization” or “intentional jitter” to move the image slightly over the sensor for the second exposure. This, with a sensor fixed in its position, also gives a shift to the second exposure so the gaps between the facets from the first exposure can be detected, and, the missing imagery recorded and spliced into the final image.
In one example shown in
This type of connection, like the coiled wire connection, make an oscillating sensor connection durable.
In summary, the advantages offered by the present invention include, but are not limited to:
High resolution digital zoom
Faster Lighter CheaperLonger focusing ranges
More reliable
Lower chromatic aberration
More accurate pixel resolution
Eliminate need for flash or floodlights
Zooming from wide angle to telephoto
A mobile communication device including a camera 150 having many of the preferred features of the present invention will now be described with reference to
It will be understood that numerous conventional features such as a battery, shutter release, aperture monitor and monitor screen have been omitted for the purposes of clarity.
The camera comprises an hermetically-sealed enclosure 154 accommodating a generally curved sensor 160 and a lens 156. Enclosure 154 is filled with Argon, Xenon or Krypton. A front view of the sensor 160 is illustrated schematically in
The center square 170 has the highest density of pixels; note that this square alone is used in the capture of telephoto images. The squares of inner ring 172 have medium density pixilation, which for normal photography gives reasonable definition. The outer ring 176 of sixteen squares has the least dense pixel count.
In this embodiment, the gaps 164 between the elements 162 are used as pathways for electrical connectors.
The camera 150 further comprises a lens shade extender arrangement 180 comprising a fixed, inner shade member 182, first movable shade member 184 and a second, radially outermost, movable shade member 186. When the operator is taking a wide angle photograph, the shade members are in a retracted disposition as shown in
For a normal perspective photograph, the shade member 184 is extended so that stray light from outside of the viewing area is blocked. In this mode, a portion of the data facets 172 of the curved sensor are compressed. To reduce processing time and storage requirements, the data from the most center area 170, with higher density of pixels, can be normalized across the entire image field.
When the user zooms out digitally to a telephoto perspective, shade member 186 is extended. In this mode, only the center portion 170 of the curved sensor 160 is used. Since only that sensor center is densely covered with pixels, the image definition will be crisp.
In operation, camera 150 uses two exposures to fill in any gaps within the sensors range, i.e., to obtain the pixel data missing from a single exposure due to the presence of gaps 164. For this purpose, the camera deploys one of two methods. In the first, as previously described, the sensor moves and a second exposure is taken in rapid succession. The processing software detects the image data that was missed in the first exposure, due to the sensor's gaps, and “stitches” that missing data into the first exposure. This creates a complete image. The process is run continuously for motion pictures, with the third exposure selecting missing data from either the preceding or the following exposure, again to create a complete image.
In the second method, a radical change to the now-standard process known in the industry as “image stabilization” is used. For the first exposure, the image is stabilized. Once recorded, this “image stabilization” is turned off, the image is shifted by the stabilization system, and the second image is taken while it is re-stabilized. In this method, a complete image is again created, but without any motion required of the sensor.
The dashed lines shown in
In another embodiment of the invention that includes intentional jittering, the lens does not move back and forth, but, rather, tilts to alter the position of the image on the sensor.
The above-described camera 150 has numerous advantages. The sealing of the enclosure 154 with a gas like argon prevents oxidation of the parts and provides thermal insulation for operation throughout a broader range of temperature.
Although the center square 170 with a high pixel density, which is relatively more expensive, it is also relatively small, and it is only necessary to provide a single such square, this keeping down the overall cost. A huge cost advantage is that it provides an acceptable digital zoom without the need for accessory lenses. Accessory lenses cost far, far more than this sensor, and are big, heavy and slow. The outer ring 176 has the smallest squares and the lowest pixel count and so they are relatively inexpensive. Thus, taking into account the entire assembly of squares, the total cost of the sensor is low, bearing in mind it is capable of providing an acceptable performance over a wide range of perspectives.
Numerous modifications may be made to the camera 150. For example, instead of being monolithic, lens 156 may comprise a plurality of elements.
The enclosure 154 is sealed with another inert gas, or a non-reactive gas such as Nitrogen, Krypton, Xenon or Argon; or it may not be sealed at all.
The pixels or facets 170, 172, 176 may be rectangular, hexagonal or of any other suitable shape. Although a central pixel and two surrounding “square rings” of pixels are described, the sensor may comprise any desired number of rings.
In
The features and modifications of the various embodiments described may be combined or interchanged as desired.
IV. Mobile Communicator with a Curved Sensor CameraThis alternative method uses multiple rapid exposures with the image moved slightly and precisely for each exposure.
In the illustrated example, four exposures are taken of the same scene, with the image shifted by ½ pixel in each of four directions for each exposure. (In practice, three, four, five or more exposures might be used with variations on the amount of image shifting used.)
For this example,
(Cameras are currently available at retail with 25 Megapixel resolution, so this tree image represents less than one millionth of the image area and would be undetectable by the human eye without extreme enlargement.)
With
This results in a data collection for this image as shown by
This third exposure, then, is represented by data collected as shown in
Now the camera has four views of the same tree image.
Current image stabilization neutralizes tiny hand tremors and even some motor or other vibrations during a single exposure, eliminating blur. That capability suggests moving the image to second, third and fourth or more positions can occur quickly.
Pixel response times are also improving regularly, to the point that digital cameras that were formerly only still cameras, have, for the most part, also become motion picture cameras in subsequent model enhancements. This also suggests that rapid multiple exposures can be done; particularly since this is the essence of motion photography.
What has not been done or suggested is changing the mode of the image stabilization mechanism so that it moves the image slightly, and by a controlled amount, for each of the multiple exposures, while stabilizing the image during each exposure.
Alternatively, moving the sensor slightly for the same effect is also a novel method.
Software interprets the four captured images and are part of this invention's claims. The software “looks” at
The software looks at all four figures and determine that whatever this is, it has a base that's above that stub, and that base is wider than the rest of the image, going three pixels horizontally. This comes from line five in
The software looks at lines three and four in
But, the software also looks at lines three in
The software averages these different conclusions and make the second tier 1 ½ A pixels tall.
The software looks at line two in all four images and realize that there is a narrower yet image atop the second tier. This image is consistently one pixel wide and one pixel high, sits atop the second tier but is always centered over the widest bottom tier, and the stub when the stub appears.
The tree is captured digitally four times.
One embodiment of this new method proposes to create a concave mold to shape the silicon after heating the wafer to a nearly molten state. Gravity then settles the silicon into the mold. In all of these methods, the mold or molds could be chilled to maintain the original thickness uniformly by reducing the temperature quickly. Centrifuging is a second possible method. The third is air pressure relieved by porosity in the mold. A fourth is steam, raised in temperature by pressure and/or a liquid used with a very high boiling point. The fourth is simply pressing a convex mold onto the wafer, forcing it into the concave mold, but again, doing so after raising the temperature of the silicon.
Heating can occur in several ways. Conventional “baking” is one. Selecting a radiation frequency that affects the silicon significantly more than any of the other materials is a second method. To enhance that second method, a lampblack-like material that absorbs most of the radiation might be placed on the side of the silicon that's to become convex, and is removed later. It absorbs the radiation, possibly burns off in the process but heats the thickness of the wafer unevenly, warming the convex side the most, which is where the most stretching occurs. A third method might be to put this radiation absorbing material on both surfaces, so the concave side, which absorbs compression tension and the convex side, which is pulled by tensile stresses, are each heated to manage these changes without fracturing.
A final method is simply machining, polishing or laser etching away the excess material to create the curved sensor.
In the this embodiment, the curved surface is machined out of the silicon or other ingot material. The ingot would be thicker than ordinary wafers. Machining could be mechanical, by laser, ions or other methods.
In the second embodiment, the wafer material is placed over a pattern of concave discs. Flash heating lets the material drop into the concave shape. This may be simply gravity induced, or, in another embodiment, may be centrifuged. Another enhancement may be to “paint” the backside with a specific material that absorbs a certain frequency of radiation to heat the backside of the silicon or other material while transmitting less heat to the middle of the sensor. This gives the silicon or other material the most flexibility across the side being stretched to fit the mold while the middle, is less heated, holding the sensor together and not being compressed or stretched, but only bent. In another embodiment, the frontside is “painted” and irradiated, to allow that portion to compress without fracturing. In another embodiment, both sides are heated at the same time, just before reforming.
Radiation frequency and the absorbent “paint” would be selected to minimize or eliminate any effect on the dopants if already inserted.
VII. Improving Image DetailsIn another embodiment of the invention, a generally constant motion is deliberately imparted to a sensor and/or an optical element while multiple exposures are taken. In another embodiment, this motion may be intermittent. Software then processes the multiple exposures to provide an enhanced image that offers greater definition and edge detail. The software takes as many exposures as the user may predetermine.
In this embodiment, the sensor is arrayed with pixels having a variable density, with the highest density in the center of the sensors. When the sensor rotates, the motion on the outer edges is far greater than at the center. Taking pictures with less than a pixel diameter of motion results in enhanced detail that is captured in the composite image. The pixels at the outer edges, where they are least densely placed, will be the largest individual pixels. The center pixels, where concentration is greatest, will have the smallest pixels. In between those extremes, the sizes will gradually change and grow as the distance from the center increases. In this way, for a fraction of a degree in rotation, the same amount of pixel change across the image takes place, and, image definition can be enhanced in this rotating direction. When a second exposure is taken with a fraction of a pixel's rotation, more edge detail of the image is captured and enhanced.
Fixed Sensor with Moving Image
In one alternative embodiment of the invention, a stationary flat or curved sensor may be used to collect data or to produce an image using an image which moves in a circular motion. In one implementation of this embodiment, the circular path of the image has a diameter which is generally less than the width of a pixel on the sensor. In one embodiment, the circular path has a diameter which is half the width of a pixel. In this embodiment, pixel density is constant across the sensor. If the image was a picture of a clock, it would move constantly in a small circle, with the number 12 always on top and the number 6 always on the bottom.
Moving Sensor with Stationary Image
In yet another alternative embodiment of the invention, a flat or curved sensor which generally constantly moves in a tight circle may be used to collect data or to produce an image. In one implementation of this embodiment, the circular path of the moving sensor has a diameter which is generally less than the width of a pixel on the sensor. In one embodiment, the circular path has a diameter which is half the width of a pixel. In other embodiments, the circular paths might have diameters of four and one half pixels, or six and one quarter pixels, or some other suitable diameter. The invention may be implemented using any movement that results in capturing added fractional images to provide detail beyond the normal pixel count's ability to detect.
The advantages of these embodiments include:
Elimination of any reciprocal movement
No vibration
No energy loss from stop and go motions
In alternative embodiments, the sensor depicted in
This embodiment enhances detail in an image beyond a sensor's pixel count, and may be used in combination with the method described in Section V, above, “Method to Capture More Detail from a Scene than the Sensor is Otherwise Capable of Recording.”
While pixel density is increasing on sensors rapidly, when pixels are reduced in size such that each pixel can sense only a single photon, the limit of pixel density has been reached. Sensitivity is reduced as pixels become smaller.
This embodiment may be utilized in combination with methods and apparatus for sensor connections described in Co-Pending U.S. patent application Ser. No. 12/655,819, filed on 6 Jan. 2010, U.S. Patent Publication No. US2010/0260494; See, especially, Paragraphs 101-113.
In yet another embodiment, miniature radios may be used to connect the output of the sensor to a micro-processor.
VIII. Alternative Embodiment Sensor Formed from Silicon Fiber FabricIn yet another embodiment of the invention, a sensor is formed from a fabric woven from silicon fibers. While this embodiment employs silicon fibers, the invention may be practiced using any light transmissive fiber material that is suitable for this purpose.
After the fabric 402 is woven, it is placed over a heated first mandrel 404, as shown in
In an alternative embodiment, a group of parallel fibers 410 is be formed into a single unwoven laminate 412, as shown in
In another variation, two sets of fibers may be laid at right angles 414, and then heated to form the fused laminate 416, as shown in
In yet another embodiment of the invention, a curved sensor 418 is formed from a number of generally flat “mini-sensors” which are positioned to form a nearly curved surface. In this embodiment, a slightly different manufacturing process is employed for each successive row of mini-sensors. In alternative embodiments, the sensor may be a conventional planar sensor.
In this embodiment, each mini-sensor is slightly smaller than the adjoining inner row so their corners do not overlap as they tilt inward to create the curve. This creates gaps which were explained in previous portions of this Specification. These gaps also provide shorter average connections to the pixels, so less space is wasted. Since either the image or the sensor itself shift and take double exposures, as explained above, the entire surface captures imagery, creating a virtual 100% “fill factor.” “Fill factor” is the industry phrase that measures the percentage area of a sensor that actually captures photons and converts them to signal data. In this embodiment, each has different pixel densities. The mini-sensor is the center of the entire sensor is the most dense, so digital zoom becomes palatable and crisp without undue expense. The next row out is less dense, saving cost, and the same amount of detail which is determined by the total pixel count is captured for normal perspective shots. The outside row (this example assumes three rows for simplicity) could is the least detailed, and, being the most numerous, is the cheapest to manufacture.
For just one of many possible examples, assume there are three rows of “mini-sensors” used to create the overall “nearly curved sensor.” The center sensor, which would be used alone for telephoto photography during digital zooming, could contain 20 MP of sensors. This creates a superior telephoto image over optical zoom telephoto images. As a small, but densely pixilated “mini-sensor,” its cost is competitive. Costs will be high due to the extremely dense pixilation. But that is counteracted by the small size required for this single “mini-sensor”, so many more are created per wafer. Assume the second row surrounding this center sensor contain eight slightly smaller 2.5 MP “mini-sensors.”
While the center “mini-sensor” is used for the maximum telephoto digital zoom imagery, when combined with the second surrounding row of “mini-sensors” the perspective changes to a normal photograph. Since the eight surrounding “mini-sensors” combine to create a 20 MP image, the center sensor is, in this embodiment, compressed down to 2.5 MP itself. This means the normal perspective photograph would be made up of 22.5 MP. Again, this is a detailed image. By compressing the center “mini-sensor” data, processing is faster and storage requirements are reduced. The massive detail from that center sensor is not needed when the image captured backs away from telephoto to a normal perspective.
Likewise, for an embodiment designed for wide angle photos, the outer row of sixteen “mini-sensors” are 1 MP each. But, since the inner rows and center “mini-sensor” are delivering unnecessary data amounts, they are all compressed to 1 MP each. This delivers a wide angle image of 25 MP, more than the human eye can resolve at an 8″×10″ enlargement. (There are 25 “mini-sensors” each delivering 1 MP of data.)
This configuration provides cost savings, as well as optimal mega-pixel counts for all three perspectives: telephoto, normal and wide angle, and for any perspective in between.
In yet another embodiment, the relatively “sparse” mini-sensors in the outermost ring are formed so that they are much larger than the other mini-sensors that are nearer to the center of the sensor, as shown in
In yet another embodiment, this sensor may be generally constantly rotated, as shown in
Yet another embodiment is shown in
Although the present invention has been described in detail with reference to one or more preferred embodiments, persons possessing ordinary skill in the art to which this invention pertains will appreciate that various modifications and enhancements may be made without departing from the spirit and scope of the Claims that follow. The various alternatives for providing a Curved Sensor Formed From Silicon Fibers that have been disclosed above are intended to educate the reader about preferred embodiments of the invention, and are not intended to constrain the limits of the invention or the scope of Claims.
LIST OF REFERENCE CHARACTERS
- 10 Camera with curved sensor
- 12 Curved sensor
- 14 Enclosure
- 16 Objective lens
- 18 Incoming light
- 20 Electrical output from sensor
- 22 Signal processor
- 24 User controls
- 26 Battery
- 28 Memory
- 30 Camera output
- 32 Facet
- 34 Gap between facets
- 36 Via
- 38 Wiring backplane
- 40 Curved sensor formed from adjoining petal-shaped segments
- 42 Petal-shaped segment
- 43a First Mandrel
- 43b Substrate
- 43c First sheet of deformable material
- 43d Dome portion of deformable material over mandrel
- 43e Hemispherical base for curved sensor
- 43f Second sheet of deformable material
- 43g Second mandrel
- 43h Ports
- 43i Empty region
- 43j Heater
- 43k Hemispherical base for curved sensor
- 44 Camera monitor
- 46 Conventional sensor with generally uniform pixel density
- 48 Sensor with higher pixel density toward center
- 50 Pixel
- 52 Shade retracted
- 54 Shade extended
- 56 Multi-lens camera assembly
- 58 Objective lens
- 60 Mirrored camera/lens combination
- 62 Primary objective lens
- 64 Secondary objective lens
- 66 First sensor
- 68 Second sensor
- 70 Mirror
- 72 Side-mounted sensor
- 74 Sensor in original position
- 76 Sensor in rotated position
- 78 Sensor in original position
- 80 Sensor in displaced position
- 82 Alternative embodiment of sensor
- 84 Alternative embodiment of sensor
- 86 Alternative embodiment of sensor
- 88 Alternative embodiment of sensor
- 90 View of rear of one embodiment of sensor
- 92 Spiral-shaped conductor
- 94 Connection to sensor
- 96 Connection to processor
- 98 View of rear of one embodiment of sensor
- 100 Accordion-shaped conductor
- 102 Connection to sensor
- 104 Connection to processor
- 106 View of rear of one embodiment of sensor
- 108 Radial conductor
- 110 Brush
- 112 Brush contact point
- 114 Annular ring
- 116 Center of sensor, connection point to processor
- 118 Schematic view of wireless connection
- 120 Transmitter
- 122 Receiver
- 124 Processor
- 150 Camera
- 154 Enclosure
- 156 Lens
- 160 Sensor
- 162 Facets
- 164 Gaps
- 170 Center square
- 172 Ring of squares
- 176 Ring of squares
- 180 Shade extender arrangement
- 182 Inner shade member
- 184 Movable shade member
- 186 Outer, movable shade members
- 190 Lens moving mechanism
- 200 Image sequence processor
- 202 Sensor capture device
- 204 Auto device
- 206 Pixel density normalization device
- 208 Image processing engine
- 210 Display/LCD controller
- 212 Compression and storage controller
- 250 Camera
- 256 Lens
- 260 Sensor
- 270 Central region facet
- 272 Surrounding region facets
- 274 Shutter control
- 280 Lens shade actuator
- 290 Focus/stabilization actuator
- 292 Lens moving
- 300 First embodiment of combined device
- 300a First embodiment of combined device
- 300b First embodiment of combined device
- 302 Housing
- 304 Micro-controller
- 305a Front side
- 305b Back side
- 306 Display screen
- 308a Touch screen interface
- 308b User interface
- 310 Terminal for power and/or data
- 314 Speaker
- 315 Antenna
- 330 View of alternative embodiment
- 334 View of alternative embodiment
- 338 View of alternative embodiment
- 340 View of alternative embodiment
- 342 Schematic illustration of moving lens with fixed flat sensor
- 344 Moving lens
- 346 Fixed flat sensor
- 348 Light path
- 350 Overhead view of
FIG. 51 - 352 Schematic illustration of moving lens with fixed curved sensor
- 354 Fixed curved sensor
- 356 Overhead view of
FIG. 53 - 358 Schematic illustration of fixed lens with moving flat sensor
- 360 Moving flat sensor
- 362 Fixed lens
- 364 Overhead view of
FIG. 55 - 365 Schematic depiction of components that impart circular motion to sensor
- 366 Spinning disc
- 367 Connecting post
- 368 Attachment point
- 370 Electric motor
- 372 Axis of motor
- 373 Perspective view of
FIG. 57 - 374 Schematic view of fixed lens over moving curved sensor
- 376 Moving curved sensor
- 377 Overhead view of
FIG. 59 - 378 Schematic illustration of components for imparting motion to lens
- 380 Band
- 382 Springs
- 384 Springs connected to cams
- 386 First cam
- 388 Second cam
- 390 First electric motor
- 392 Second electric motor
- 394 Series of nine views of rotating sensor
- 396 Sensor
- 398 Pixels
- 400 Optical fiber
- 402 Woven mesh of optical fibers
- 404 Heated first mandrel
- 405 Upper mandrel
- 406 Fabric dome
- 408 Trim excess fabric
- 410 Parallel fibers
- 412 Fused laminate
- 414 Two sets of fibers at right angles
- 418 Sensor with rows of mini-sensors of increasing size
- 420 Sensor
- 422 First ring of mini-sensors
- 424 Second ring of mini-sensors
- 426 Outermost ring of mini-sensors
- 428 Rotating sensor with rows of mini-sensors of increasing size
- 430 Sensor with rows of mini-sensors with every other mini-sensor shifted upwards
Claims
1. A method for making a sensor, comprising the steps of:
- providing a plurality of light transmissive fibers;
- arranging said plurality of light transmissive fibers into a first group and a second group;
- configuring said first group of said plurality of light transmissive fibers so that they are all generally parallel to each other;
- said first group and said second group of said plurality of light transmissive fibers being configured so that they are generally orthogonal to each other;
- weaving said second group of said plurality of light transmissive fibers into said first group of said plurality of light transmissive fibers into a woven fabric;
- placing said woven fabric over a first heated mandrel;
- compressing said woven fabric on said first heated mandrel using a second heated mandrel;
- fusing both of said first and said second groups of said plurality of light transmissive fibers together to form a woven fabric; and
- trimming away excess fabric.
2. A method as recited in claim 1, in which said plurality of light transmissive fibers are formed from silicon.
3. A method as recited in claim 1, in which said both of said first and said second mandrels are curved so that a curved woven fabric is produced.
4. A method for making a sensor, comprising the steps of:
- providing a plurality of light transmissive fibers;
- arranging said plurality of light transmissive fibers so that they are generally parallel to each other;
- heating said plurality of light transmissive fibers to form an unwoven laminate; said unwoven laminate being generally flat; and
- heating said unwoven laminate on a mold.
5. A method as recited in claim 4, in which said plurality of light transmissive fibers is formed from silicon.
6. A method as recited in claim 4, in which said mold is curved to produce a curved unwoven laminate.
7. A method comprising the steps of:
- providing a first set and a second set of light transmissive fibers;
- arranging said first and said second set of light transmissive fibers at right angles;
- heating said first and said second set of light transmissive fibers to form a fused laminate; and
- forming said fused laminate into a sensor.
8. A method as recited in claim 7, in which said first and said second sets of light transmissive fibers are formed from silicon.
9. A method as recited in claim 7, in which said fused laminate in formed into a curved sensor.
10. A method comprising the steps of:
- providing a sensor;
- providing a plurality of generally flat mini-sensors; and
- affixing said plurality of generally flat mini-sensors to said sensor so that said generally flat mini-sensors are arranged in horizontal rows, and so that each successive row is less densely populated with said generally flat mini-sensors.
11. A method as recited in claim 10, in which two edges of said mini-sensors are generally parallel to the base of said sensor.
12. A method as recited in claim 10, in which every other mini-sensor in each row of mini-sensors is positioned slightly higher than its two neighboring mini-sensors in the same row.
13. A method as recited in claim 10, in which said sensor is generally continuously rotated about its vertical axis.
14. A method as recited in claim 10, in which said sensor is a generally curved sensor.
Type: Application
Filed: Jun 30, 2011
Publication Date: Jun 28, 2012
Inventor: Gary Edwin Sutton (La Jolla, CA)
Application Number: 13/135,402
International Classification: C03B 37/15 (20060101); B32B 38/00 (20060101); B23P 11/00 (20060101); B32B 37/14 (20060101); G02B 6/255 (20060101); G03B 13/00 (20060101);