Abstract: A method and system for converting a patterned color image in a first color space to a second color space, prior to image compression is provided. The process includes, dividing the image into plural tiles; converting the plural tiles from the first color space to the second color space; retaining certain pixel values in the second color space; discarding certain pixel values in the second color space; and creating a pattern in the second color space, similar to the patterned color image in the first color space. Also, a method for decoding a compressed image pattern from a first color space to second color space is provided. The method includes, assembling pixels from a first pattern based on the first color space; transforming triplets form the first color space to the second color space; and determining pixel values for a dominant color.

Abstract: A method and system for automatic bad pixel correction in image sensors is provided. The process includes identifying outlier pixels, identifying bad pixels, and performing bad pixel correction. Bad pixels are identified by comparing pixels in a single row or more than one row. A bad pixel value is replaced by a pixel value that depends on the pixel value of non-bad pixels located next to the bad pixel. The system includes means for identifying outlier pixels, means for identifying bad pixels, and means for performing bad pixel correction.

Abstract: An automated license plate locator and reader (10) which provides novel methods for 1) correcting for perspective distortion (16), 2) locating the license plate (20), 3) reading the license plate (22) and 4) improving the confidence rating of the output signal (228). The novel method for accommodating a wide range of viewing geometries involves warping the image to eliminate perspective distortion of the plate resulting from roadside camera placement. The novel method for locating plates involves correlating an "idealized" edge template of a license plate with an edge-enhanced version of the image. Edge enhancement and shadow reduction is achieved by applying a Laplacian operator over the input image. Performance is further improved by filtering out distracting edges within the resulting correlation surface and then locating the highest peaks within this filtered correlation surface. The location of the highest peaks provides the location of the plate.

Abstract: A vehicular traffic monitoring system incorporates an array of photosensors and a nonlinear resistive network for identifying, locating, and processing outliers in sensor images of a highway or intersection. The camera system can be mounted on a pole or overpass to provide an image of the roadway or intersection. Areas of the outlier network ("video loops") are designated to correspond to selected areas of the roadway. Images are received by the outlier detection network with all data path switches closed between sensor elements and their corresponding network nodes. The system detects the presence of objects in the image by comparing the brightness or intensity of each pixel with that of the background. If the intensity of a pixel is significantly different from the background level, the data path switch corresponding to that pixel is opened. A readout of the state of all the switches in the network yields a map of outlier points for each video frame.

Abstract: Image data is processed by a low level feature detection processor that extracts low level features from an image. This is accomplished by converting a matrix of image data into a matrix of orthogonal icons that symbolically represent the image scene using a predetermined set of attributes. The orthogonal icons serve as the basis of processing by means of a high level graph matching processor which employs symbolic scene segmentation, description, and recognition processing that is performed subsequent to the low level feature detection. This processing generates attribute graphs representative of target objects present in the image scene. High level graph matching compares predetermined attributed reference graphs to the sensed graphs to produce a best common subgraph between the two based on the degree of similarity between the two graphs. The high level graph matching generates a recognition decision based on the value of the degree of similarity and a predetermined threshold.