Abstract: An image processing apparatus calculates estimated pixel values of respective pixels of a provisional high-resolution image by interpolation on basis of pixel values in a reference frame. Interior of the reference frame is segmentalized into an edge region, a texture region or a flat region and others on a basis of pixel values of respective pixels. The respective pixels in the reference frame are set as target pixels one by one in sequence. Corresponding positions on the provisional high-resolution image are calculated, of the respective target pixels in decimal accuracy on a basis of information on the segmented regions including the target values. The estimated pixel values are modified so that differences from the pixel values of the target pixels to provisionally estimated pixel values obtained from the estimated pixel values of the provisional high-resolution image for pixels around the corresponding positions of the target pixels, becomes smaller. Modified pixel values and then obtained.

Abstract: This invention provides a vehicle-borne system and method for traffic sign recognition that provides greater accuracy and efficiency in the location and classification of various types of traffic signs by employing rotation and scale-invariant (RSI)-based geometric pattern-matching on candidate traffic signs acquired by a vehicle-mounted forward-looking camera and applying one or more discrimination processes to the recognized sign candidates from the pattern-matching process to increase or decrease the confidence of the recognition. These discrimination processes include discrimination based upon sign color versus model sign color arrangements, discrimination based upon the pose of the sign candidate versus vehicle location and/or changes in the pose between image frames, and/or discrimination of the sign candidate versus stored models of fascia characteristics.

Abstract: Provided is an image-resolution-improvement apparatus and method which can increase the resolution of an input image at a high magnification to thereby obtain a high-quality final image. The apparatus includes a textured-region-detection unit to detect a texture region in an input image; and a final-image-generation unit to synthesize a first intermediate image and a second intermediate image, which are obtained by applying different interpolation techniques to the texture region and a non-texture region excluding the texture region and generating a final image.

Abstract: The invention provides an image encoding apparatus which can improve image quality of an output image while further reduce the amount of attribute. A determination unit determines an area including a character/line drawing as a foreground image area based on an input multi-valued image. A foreground image generator generates foreground image in binary representation so that a first encoder performs MMR encoding on the foreground image. A background image generator generates multi-valued background image data by replacing the value of a multi-valued pixel in a position of the character/line drawing in the foreground image area with a replacement value calculated from the pixel values in a position of the non-character/line drawing pixel. A second encoder performs JPEG encoding on the background image. A mask unit masks attribute for pixels within the foreground image area with a predetermined value to output the masked data to a third encoder.

Abstract: Systems and methods for transprojection of geometry data acquired by a coordinate measuring machine (CMM). The CMM acquires geometry data corresponding to 3D coordinate measurements collected by a measuring probe that are transformed into scaled 2D data that is transprojected upon various digital object image views captured by a camera. The transprojection process can utilize stored image and coordinate information or perform live transprojection viewing capabilities in both still image and video modes.

Abstract: The invention is generally related to the systems and methods through which household appliances can provide intuitive and enhanced interactivity through visual inputs. In particular, the disclosure presents methods relating to control of a household appliance, such as a microwave or a refrigerator, via recognition of visual indicia, such as graphics, text, and the like, using a visual sensor such as a camera.

Abstract: The spatial location and azimuth of an object are computed from the locations, in a single camera image, of exactly two points on the object and information about an orientation of the object. One or more groups of four or more collinear markers are located in an image, and for each group, first and second outer markers are determined, the distances from each outer marker to the nearest marker in the same group are compared, and the outer marker with a closer nearest marker is identified as the first outer marker. Based on known distances between the outer markers and the marker nearest the first outer marker, an amount of perspective distortion of the group of markers in the image is estimated. Based on the perspective distortion, relative distances from each other point in the group to one of the outer markers are determined. Based on the relative distances, the group is identified.

Abstract: A system creates three-dimensional computer models of physical objects by displaying illumination patterns on a display device to incidentally illuminate a physical object. A video camera acquires images of the object illuminated by the patterns. The patterns can include motion and multiple colors for acquiring images with large variations in surface shading of the object. Shading values from acquired images of the object are analyzed to determine the orientations of points on the object's surface. The system evaluates the quality of acquired images and selects patterns tailored to specific attributes of objects. The points' orientations are determined by comparing the points' shading values with an illumination model or shading values acquired from a calibration shading object. A model surface is fitted to the points' orientations. Applications may utilize the model for any purpose, including creating and exchanging customized virtual objects, enhanced object tracking, and videoconferencing applications.

Abstract: An image processing system interpolates image data of an image array by ascertaining shift invariant points and non-shift invariant points within the array. The average illumination and the second order derivative are determined for the shift invariant locations. The second order derivative and the intensity at the non-shift invariant locations for each of the non-shift invariant points are estimated. The color data for each color element is determined from the image data and second order derivative. The second order derivative is multiplied by a scaling factor for selectively smoothing and sharpening the second order derivative. The color data values of adjacent color element to enhance a resolution of the image data.

Abstract: A plurality of iris images are acquired (SA0), and aggregation of iris images of which distribution of pupil openings is uniform is acquired from the plurality of iris images by duplication and/or deletion (SA1). Features are generated from the respective iris images that belong to the aggregation (SA2), and a predetermined number of registration features are selected from the features, using authentication performance as an evaluation index.

Abstract: The disclosure is directed to techniques for region-of-interest (ROI) video processing based on low-complexity automatic ROI detection within video frames of video sequences. The low-complexity automatic ROI detection may be based on characteristics of video sensors within video communication devices. In other cases, the low-complexity automatic ROI detection may be based on motion information for a video frame and a different video frame of the video sequence. The disclosed techniques include a video processing technique capable of tuning and enhancing video sensor calibration, camera processing, ROI detection, and ROI video processing within a video communication device based on characteristics of a specific video sensor. The disclosed techniques also include a sensor-based ROI detection technique that uses video sensor statistics and camera processing side-information to improve ROI detection accuracy.

Abstract: After pre-printed cards are manufactured at least a corner of each card is imaged. It is then determined whether the position and orientation of markings in the image with respect to the corner are within tolerance. If not, the card is rejected. If the markings are, or include, indicia, these indicia may be identified to allow identification of the card. This allows a determination of whether the card is in an intended order, or it allows selection of the batch into which the card is placed. It also means that cards in a given set of cards arranged in a first order can be randomized by first buffering cards in buffers and then outputting the cards to an output conveyor.

Abstract: A system and method for character recognition with document orientation determination is shown. The method is a detection of simple page orientation based on a limited version of character recognition. The method includes binairizing an input image which has a plurality of alphanumeric characters with a first orientation. The method continues with extracting the connected components and determining a second orientation where the second orientation is based on a 90° turn clockwise or counterclockwise or, in the alternative, no turn from the first orientation. The second orientation will result in a 180° variance from the proper orientation or it will be the proper orientation. The method continues with implementing a limited version of optical character recognition for an analysis of a character and determining if that second orientation is upside down, based at least in part on the analysis. This method generally uses the character “i” for analysis.

Abstract: A method of extracting a feature from a point cloud comprises receiving a three-dimensional (3-D) point cloud representing objects in a scene, the 3-D point cloud containing a plurality of data points; generating a plurality of hypothetical features based on data points in the 3-D point cloud, wherein the data points corresponding to each hypothetical feature are inlier data points for the respective hypothetical feature; and selecting the hypothetical feature having the most inlier data points as representative of an object in the scene.

Abstract: The invention relates to a method for the joint of registering, archiving, contrast-retaining merging, and visualizing of 2D x-ray images and reconstructed 2D projections or 3D views of generated image data in minimally invasive medical interventions or operations performed under x-ray control on internal organs, areas of tissue, lesions, or pathological structures in a patient's body. A first image and the image data of a second image being mixed over the first image are mutually linked, co-registered, and merged using an alpha-channel-based mixing technique, whereby the color values or gray-scale values of the individual pixels of the first image are prior to being mixed-over with the segmented foreground of the second image brightened through the application of a color-value or gray-scale-value for compensating or lessening the reduction in contrast that occurs during mix-over in the first image before being displayed graphically on a monitor terminal.

Abstract: Provided are methods and apparatuses for converting a lower bit-plane image, and methods and apparatuses for inverse-converting the lower bit-plane image, which increase the correlation of the lower bit-plane image. According to the method of converting a lower bit-plane image, pixels of the lower bit-plane image are sequentially added or subtracted so as to generate a converted lower bit-plane image. Accordingly, the lower bit-plane image has higher regularity, and thus the efficiency of image processing using the lower bit-plane image is improved.

Abstract: In view of the fact that the reproducibility of a finger vein pattern is deteriorated in an image of the finger vein due to external scattered light, displacement, change in the amount of blood flow and the like, a multiple-valued pattern matching method is adopted so that such changes can be dealt with by converting a data map itself into a multiple-valued map.

Abstract: An image processing system is disclosed that includes a buffer unit configured to store a predetermined pixel and peripheral pixels neighboring the predetermined pixel in image data in a first direction and a second direction perpendicular to the first direction; and a first edge-detecting circuit configured to calculate a maximum value and a minimum value of a set of pixels selected from the peripheral pixels that have the same color as the predetermined pixel to calculate a difference value between the maximum value and the minimum value.

Abstract: A histogram has ranges, a value for each range, and a time value. A pattern is generated, having a zero value for each range having a zero value, and a one value for each range having a non-zero value. Any non-zero values of the ranges are compressed into a compressed value. The pattern and the time value match or do not match any given entry of a pattern table having a number of entries, each of which has an index value, a time range, and a pattern. Where the pattern and the time value do not match any given entry of the pattern table, a new entry is added to the pattern table, and is considered the given entry. An entry is then added to a histogram table. This entry has the time value, the index value of the given entry of the pattern table, and the compressed value.

Abstract: The present invention is a method and system for automatically analyzing a category in a plurality of the categories in a physical space based on the visual characterization, such as behavior analysis or segmentation, of the persons with regard to the category. The present invention captures a plurality of input images of the persons in the category by a plurality of means for capturing images. The present invention processes the plurality of input images in order to understand the shopping behavior of the persons with the sub-categories of the category and analyzes the level of engagement and decision process at the sub-category level. The processes are based on a novel usage of a plurality of computer vision technologies to analyze the visual characterization of the persons from the plurality of input images. The physical space may be a retail space, and the persons may be customers in the retail space.