Abstract: Techniques of designing a sensing system for pseudo 3D mapping in robotic applications are described. According to one aspect of the present invention, an image system is designed to include at least two linear sensors, where these two linear sensors are positioned or disposed orthogonally. In one embodiment, the two linear sensors are a horizontal sensor and a vertical sensor. The horizontal sensor is used for the lidar application while the vertical sensor is provided to take videos, namely scanning the environment wherever the horizontal sensor misses. As a result, the videos can be analyzed to detect anything below or above a blind height in conjunction with the detected distance by the lidar.

Abstract: Designs of a sensor module with fewer number of pins are described. According to one aspect of the designs, a sensor module operates on a predefined number of clocks in a clock signal in contrast of rising or falling edges of clocks or pulses thereof commonly relied upon in a prior art sensor module, thus reducing considerably the requirements on the clock signals needed to support the sensor module. More importantly, the pins of the sensor module are far fewer than those in a prior art system. Subsequently, the complexity of supporting circuits for the sensor module in the sensor module of the present invention can be simplified.

Abstract: Designs of an anti-blooming sensing element or simply sensor are described. According to one aspect of the designs, each sensing element includes a photosensor, a pair of first and second circuits, a store device and a readout circuit. The first and second circuits, sandwiching the photosensor from a circuit perspective, are mirrored and balanced in impedance. The first circuit provides an inherent mechanism to discharge excessive charge accumulated on the photosensor before a predefined exposure time ends. Each of the two circuits includes two transistors.

Abstract: Techniques of designing a sensing system for pseudo 3D mapping in robotic applications are described. According to one aspect of the present invention, an image system is designed to include at least two linear sensors, where these two linear sensors are positioned or disposed orthogonally. In one embodiment, the two linear sensors are a horizontal sensor and a vertical sensor. The horizontal sensor is used for the lidar application while the vertical sensor is provided to take videos, namely scanning the environment wherever the horizontal sensor misses. As a result, the videos can be analyzed to detect anything below or above a blind height in conjunction with the detected distance by the lidar.

Abstract: A focusing mechanism or module is designed to reduce the size of an optical lens-based focusing system that would be otherwise used in a portable device. According to one aspect of the present invention, the focusing mechanism includes a light guide with first and second sides. The light guide includes a plurality of light passages slanted inwardly formed evenly from the first side towards a center of the second side, wherein the light guide, disposed on top of the image sensor, collects a reflected light from a human body part and focuses the reflected light onto the image sensor, each of photosensors generates an proportional charge from the reflected light.

Abstract: A focusing mechanism or module is designed to reduce the size of an optical lens-based focusing system that would be otherwise used in a portable device. According to one aspect of the present invention, the focusing mechanism includes a light guide with first and second sides. The light guide includes a plurality of light passages slanted inwardly formed evenly from the first side towards a center of the second side, wherein the light guide, disposed on top of the image sensor, collects a reflected light from a human body part and focuses the reflected light onto the image sensor, each of photosensors generates an proportional charge from the reflected light.

Abstract: Techniques to monitor multiple targets with a single camera are disclosed. In one embodiment, an image sensor is provided with two or more readout circuits, each operating independently and is designed to read out charges from a designated area of the image sensor. When two or more designated sensing areas in the image sensor are being focused onto different objects and read out respectively, such an image sensor is capable of monitoring multiple targets. When placed in traffic surveillance, a camera equipped with such an image sensor is able to monitor multiple forward and backward lanes in near or far field. Further with the control of the designated areas, different resolutions of the images may be produced.

Abstract: Designs of multi-band sensor array to generate multi-spectral images are disclosed. According to one aspect of the present invention, a multi-band sensor array includes one linear sensor configured to sense a scene in panchromatic spectrum to produce a panchromatic (PAN) sensing signal, and four color-band linear sensors to sense the same scene in different color bands to produce respective sensing signals. These sensors are packaged in a single module that is disposed on a single optical plane when used to scan a scene. A multi-spectral image is produced by combining these sensing signals. Further a unique packaging of the sensor array and a combination of soft and hard PCB are disclosed to withstand extremes in a harsh environment.

Abstract: Designs of multi-band sensor array to generate multi-spectral images are disclosed. According to one aspect of the present invention, a multi-band sensor array includes one linear sensor configured to sense a scene in panchromatic spectrum to produce a panchromatic (PAN) sensing signal, and four color-band linear sensors to sense the same scene in different color bands to produce respective sensing signals. These sensors are packaged in a single module that is disposed on a single optical plane when used to scan a scene. A multi-spectral image is produced by combining these sensing signals. Further a unique packaging of the sensor array and a combination of soft and hard PCB are disclosed to withstand extremes in a harsh environment.

Abstract: Designs of image sensors with subpixels are disclosed. According to one aspect of an image sensor in one embodiment, subpixels within a pixel are designed without significantly increasing the cell or pixel area of the pixel. The readouts from the subpixels are accumulated to increase the sensitivity of the pixel without increasing the area of the image sensor. According to another aspect of the image sensors in the present invention, some subpixels within a pixel are respectively coated with filters, each designed for a frequency range. Thus the frequency response of a CMOS image sensor can be enhanced significantly according to application.

Abstract: Techniques to monitor multiple targets with a single camera are disclosed. In one embodiment, an image sensor is provided with two or more readout circuits, each operating independently and is designed to read out charges from a designated area of the image sensor. When two or more designated sensing areas in the image sensor are being focused onto different objects and read out respectively, such an image sensor is capable of monitoring multiple targets. When placed in traffic surveillance, a camera equipped with such an image sensor is able to monitor multiple forward and backward lanes in near or far field. Further with the control of the designated areas, different resolutions of the images may be produced.

Abstract: A multi-pixel row wafer-scale cluster image sensor chip (WCISC) is proposed. Expressed in X-Y-Z coordinates with its pixel rows along X-axis, the WCISC converts areal image frame (IMFM) into areal image frame signal (AIFS). The WCISC includes multiple imaging pixel rows PXRW1, . . . , PXRWM. Each PXRWi has photoelectrical sensing elements spanning pixel row width PRWi and producing a pixel row image signal PRISi. Each PXRWi is offset from PXRW1 by distance XOFSTi and spaced from PXRWi?1 by distance SPi?1,I such that X- and Y-extremities of (PXRW1, . . . , PXRWM) define IMFM. The WCISC is so configured that any image pixel sweeping through IMFM will be sensed by at least one imaging pixel row. In the presence of Y-directional relative motion between WCISC and IMFM and an external electronic imaging controller (EEIC) interfacing with the WCISC, the EEIC can extract all PRISi from WCISC and reconstruct the AIFS.

Abstract: A wafer-scale linear image sensor chip (WLISC) is proposed with gapless pixel line and signal readout circuit segments. The WLISC converts pixel line image (PLI) of length LPL into line image signal (LIS). The WLISC includes a linear array of sensor segments. Each sensor segment includes a gapless local pixel line segment (LPLSj) of sensing elements. The LPLSj converts portion of the PLI into a raw image segment signal set (RISSj). The LPLSj set forms a gapless global pixel line (GPL) corresponding to PLI. The sensor segment also includes readout circuit segment (RCSj) coupled to LPLSj for processing RISSj into a readout image segment signal set (ROSSj). The RCSj has a set of contact pads (CTPj) for off-chip interconnection. Upon off-chip interconnection of the CTPj set thus composing the ROSSj set into LIS, the WLISC functions as a key part of a linear image sensor system of image length LPL.

Abstract: A long-length industry camera image sensor (LICIS) is proposed for, expressed in X-Y-Z coordinates, converting a pixel line image (PLI) of length LPL along X-direction into a line image signal (LIS). The LICIS includes a full-width linear image sensor (FLIS) of length LIS along X-direction and displaced from the PLI along Z-direction by an imaging distance DIMG for converting an incident line image (ILI) impinging upon its FLIS top surface into the LIS. Where LIS is about equal to LPL. The LICIS also has a full-width linear rod lens (FLRL) of length LRL along X-direction and displaced from the PLI in Z-direction by a working distance DWKG. Where LRL is about equal to LPL and DWKG is selected such that the PLI gets focused by the FLRL into the ILI at the FLIS top surface with an imaging magnification factor of about 1:1.

Abstract: A multi-pixel row wafer-scale cluster image sensor chip (WCISC) is proposed. Expressed in X-Y-Z coordinates with its pixel rows along X-axis, the WCISC converts areal image frame (IMFM) into areal image frame signal (AIFS). The WCISC includes multiple imaging pixel rows PXRW1, . . . , PXRWM. Each PXRWi has photoelectrical sensing elements spanning pixel row width PRWi and producing a pixel row image signal PRISi. Each PXRWi is offset from PXRW1 by distance XOFSTi and spaced from PXRWi?1 by distance SPi?1,I such that X- and Y-extremities of (PXRW1, . . . , PXRWM) define IMFM. The WCISC is so configured that any image pixel sweeping through IMFM will be sensed by at least one imaging pixel row. In the presence of Y-directional relative motion between WCISC and IMFM and an external electronic imaging controller (EEIC) interfacing with the WCISC, the EEIC can extract all PRISi from WCISC and reconstruct the AIFS.

Abstract: A time delay integration (TDI) based MOS photoelectric pixel sensing circuit (TDIPSC) is proposed. The TDIPSC includes multi-element photoelectric pixel sensor (MEPS) having sub-pixel sensor elements SPSE1, . . . , SPSEM respectively converting a portion of pixel light into sub-pixel photoelectric signals (SPPES1, . . . , SPPESM). The TDIPSC also includes intermediate photoelectric signal accumulators (PESA1, . . . , PESAM) where any PESAk can be switchably coupled to any SPSEj via switching transistors for receiving a corresponding SPPESj from it and accruing an accumulated photoelectric signal ACPESk. A readout circuit (ROC) switchably coupled to any PESAk serves to remove and read the ACPESk. A TDI-sequence controller (TDISC) coupled to the SPSEs, the PESAs and the ROC executes a time sequence of cyclic coupling among them. The TDISC produces, via the ROC, a final time signal SQTS equal to the time delayed summation of the (SPPES1, . . . , SPPESM) with a reduced SNR.

Abstract: A wafer-scale linear image sensor chip (WLISC) is proposed with gapless pixel line and signal readout circuit segments. The WLISC converts pixel line image (PLI) of length LPL into line image signal (LIS). The WLISC includes a linear array of sensor segments. Each sensor segment includes a gapless local pixel line segment (LPLSj) of sensing elements. The LPLSj converts portion of the PLI into a raw image segment signal set (RISSj). The LPLSj set forms a gapless global pixel line (GPL) corresponding to PLI. The sensor segment also includes readout circuit segment (RCSj) coupled to LPLSj for processing RISSj into a readout image segment signal set (ROSSj). The RCSj has a set of contact pads (CTPj) for off-chip interconnection. Upon off-chip interconnection of the CTPj set thus composing the ROSSj set into LIS, the WLISC functions as a key part of a linear image sensor system of image length LPL.

Abstract: An areal active pixel image sensor (AAPS) with programmable row-specific gain is disclosed for converting hyper-spectral light image into video output signal (VOS). The AAPS includes: a) An areal active pixel sensor (APS) array each capable of photoelectrically converting and integrating an incident pixel light into a photoelectric signal through an integration time period TNT with a photoelectric signal gain GPE. b) A video output signal conditioner (VOSC), coupled to the APS array, for multiplexing and amplifying the photoelectric signals into the VOS with an electric signal gain GEE. c) The VOSC further programmably sets at least one of GPE and GEE to be row-specific. Consequently, the AAPS exhibits an overall photoelectric signal gain of GOA=GPE×GEE that is row-specific and it can compensate for image signal distortion caused by non-uniform spectral response of the APS elements during hyper-spectral imaging.

Abstract: A full-width line image-sensing head (FLIH) is proposed for, expressed in X-Y-Z coordinates, converting a pixel line image (PLI) of length LPL along X-direction into a line image signal (LIS). The FLIH includes a full-width linear image sensor (FLIS) of length LIS along X-direction and displaced from the PLI along Z-direction by an imaging distance DIMG for converting an incident line image (ILI) impinging upon its FLIS top surface into the LIS. Where LIS is about equal to LPL. The FLIH also has a full-width linear rod lens (FLRL) of length LRL along X-direction and displaced from the PLI in Z-direction by a working distance DWKG. Where LRL is about equal to LPL and DWKG is selected such that the PLI gets focused by the FLRL into the ILI at the FLIS top surface with an imaging magnification factor of about 1:1.

Abstract: An areal active pixel image sensor (AAPS) with programmable row-specific gain is disclosed for converting hyper-spectral light image into video output signal (VOS). The AAPS includes: a) An areal active pixel sensor (APS) array each capable of photoelectrically converting and integrating an incident pixel light into a photoelectric signal through an integration time period TNT with a photoelectric signal gain GPE. b) A video output signal conditioner (VOSC), coupled to the APS array, for multiplexing and amplifying the photoelectric signals into the VOS with an electric signal gain GEE. c) The VOSC further programmably sets at least one of GPE and GEE to be row-specific. Consequently, the AAPS exhibits an overall photoelectric signal gain of GOA=GPE×GEE that is row-specific and it can compensate for image signal distortion caused by non-uniform spectral response of the APS elements during hyper-spectral imaging.