Abstract: Inspection of solder paste on a printed circuit board using a histogram of an image before printing (pre-application image) to normalize an image after printing of the printed circuit board (post-application image) is described. Existing lighting and optics used for alignment of the screen-printing stencil to the printed circuit board are used for the solder paste inspection.

Abstract: This relates to a classification generation method whereby, in order to efficiently and accurately recognize a pattern having a large number of characteristics, a pattern classification tree is generated, with which a macro structural characteristic of a pattern is appropriately reflected and a competitive relationship between categories is adequately reflected, and to a method for recognizing an input pattern by using the generated classification tree. When an input pattern is formed using strokes, a training stroke is divided into a plurality of segments, and vector quantization is performed for the strokes in the segments. Among the quantized strokes in the segments, adjacent stroke sets are synthesized to repetitively generate upper rank stroke vectors. A stroke vector for which a predetermined entropy function is maximized is selected from the upper rank stroke vectors in a layered stroke vector series, and development is performed extending down into the lower rank stroke vector sets.

Abstract: In order to progressively transmit a digital image (IM), a method for coding the image comprises the division (E1) of the image into vectors (V.sub.m,n), the selection (E2) of a code vector (VC.sub.d) associated with an index (A.sub.d) for each of the image vectors, and the replacement (E2) of the image vectors by the indices associated with the selected code vectors, in order to produce a plane of indices (PI) representing the image. The method comprises the steps of sub-sampling (E3) the plane of indices (PI) into sub-planes of indices (SPI.sub.1 to SPI.sub.F) respectively comprising indices selected from the plane of indices, and ordering (E4) the sub-planes of indices according to a predetermined order before transmitting them.

Abstract: A method and system for detecting defects in the design of a photolithographic mask or a printed wafer. It derives an adaptive inspection algorithm that allows for a tighter inspection of a mask to a data set which has repeatable differences. The inspection should allow flexibility to remove un-important differences while maintaining a tight inspection capability.

Abstract: DC and AC DCT transform coefficients of an INTRA coded block are differentially encoded by selecting predictor DC and AC coefficients from a left-hand neighboring block and a top neighboring block. Each block is coded according to a frame mode, a reordered field mode, and a non-reordered field mode. The AC predictor block is selected according to the respective coding modes of the blocks, and the block in which a DC predictor resides. The top block is selected as an AC predictor when the top block and current block are both reordered field mode, or both frame mode and/or non-reordered field mode, and the DC predictor resides in the top block. Zeroed AC spatial transform coefficients are used in place of the AC spatial transform coefficients from the selected block when the selected block is not INTRA coded, or does not reside in the same Video Object Plane (VOP) as the current block. DC coefficients may be non-linearly quantized.

Abstract: According to the still image encoder, as to frame data inputted from the digital video camera, odd field data is held in the odd field memory and even field data is held in the even field memory. The control section divides odd field data and even field data into predetermined blocks, calculates the absolute value total of differences between each pixel data in odd field data and each pixel data in even field data, and compares the total with the predetermined threshold. The DMA controller transfers pixel data of the block in the odd field data to the even field memory when the total is larger than the threshold. The control section codes frame data consisting of odd field data and even field data into JPEG data.

Abstract: A multiplexing optical system necessary for transforming input information into effective feature vectors at high speed and with high accuracy, without loss of a spatial frequency component of multiplexed object vector information, for example. Also presented are a feature vector transformation apparatus using the multiplexing optical system, and others. The multiplexing optical system performs a Fourier transform in parallel on multiplexed vector information lying in an input plane (R) as an object to be processed. The system includes a Fourier transform lens (30) for performing a Fourier transform in parallel, and satisfies the condition of (k.sub.R .lambda.f.sub.F /a<p) in an arbitrary cross-section containing an optical axis, where k.sub.

Abstract: A method and apparatus for adaptively selecting a coding mode for video encoding systems using the techniques of predictive coding, interpolative coding, motion compensation, discrete cosine transformation (DCT), adaptive quantization, or variable length coding. In one system, a value related to an amount of total code, which is composed of current video code and its coding parameters for each of the above coding modes, is determined, respectively, and a coding mode corresponding to the smallest value is selected. Then, the encoding is performed using the selected coding mode.

Abstract: A process for producing a compressed representation of 2D and 3D images. The image is represented in a compressed form by approximating regions with slowly changing brightness or color, by a background, formed by low degree polynomials. Fast brightness or color changes are represented by special models, including local models and curvilinear structures. Visual adjacency relations between the models are identified, the background partition represents these adjacency relations, curvilinear structures are approximated by spline functions. The three-dimensional image is represented by producing one or several compressed images of the scene from different positions, in which a depth value is associated to each model. A view of the scene from any prescribed point is produced by a geometric processing of these compressed data.

Abstract: An apparatus for vector-quantizing an image signal having a low transmission rate which includes a four-band linear filter para-unitary (LPPU) filter group for transforming an input image signal into transform coefficients which are represented by 4.times.4 frequency bands. A code modulator differential-pulse-code-modulates a direct current (DC) component among the transform coefficients output from the four-band LPPU filter group. A coefficient extractor extracts a predetermined number of low-frequency alternating current (AC) component coefficients from the transform coefficients output from the four-band LPPU filter group. A classifier determines a corresponding class, from a plurality of predetermined classes, using the low-frequency AC component coefficients extracted by the coefficient extractor, and outputs class information representing the determined class.

Abstract: The present invention discloses a super resolution pan-sharpening technique that is used to increase the resolution of a multi-spectral signal using a panchromatic signal. Typically, multi-spectral signals, i.e., signals that contain a defined spectral band, have lower resolution than panchromatic signals since the multi-spectral signals normally contain fewer photons and require larger detectors for similar exposure periods. The technique of the present invention utilizes the panchromatic signal to increase the resolution of the multi-spectral signals to provide an increased resolution color output. The resolution of other spectral signals, such as infrared signals and UV signals, or any other desired signal, can also be increased in accordance with the techniques of the present invention.

Abstract: A method, apparatus and computer program product are provided for querying by image colors using a JPEG format for images from the internet. First an image color is selected. A DCT coefficient is acquired from a JPEG image. The acquired DCT coefficient is compared with the selected image color. A match within the JPEG image is identified responsive to the acquired DCT coefficient being near the selected image color. An average color can be used for querying JPEG formatted images from the internet.

Abstract: A block-matching motion estimation apparatus in the form of linear systolic array architecture.

Abstract: A hierarchical object encoding technique or process capable of representing general binary arbitrary shapes that include, for example, embedded or disconnected components. The method decomposes successive layers of general binary arbitrary shapes into simple arbitrary shapes. Each mask formed in this manner is a simple arbitrary shape having only a continuous outer boundary. Accordingly, each outer boundary is encoded, preferably by a contour encoding method, to provide accurate encoding of general binary shapes.

Abstract: Video data representing a scene are processed to improve encoding efficiencies. The video data are segmented into rigidly and non-rigidly moving video objects. The segmentation is preformed by estimating local motion vectors for the video data of a sequence of frames. The local motion vectors are clustered to determine dominant motions, and video data having motion vectors similar to the dominant motions are segmented out as rigidly moving video objects. For these objects, motion parameters are robustly estimated. Using the motion parameters, the rigid video objects of the frames are integrated in one or more corresponding sprites stored in a long-term memory. The sprites can be used in a two-way motion compensated prediction technique for encoding video data where blocks of video data are encoded either from sprites based on rigid motion parameters, or from a previous frame based on local motion parameters.

Abstract: A motion vector detecting apparatus for detecting a motion vector at a below-pixel precision of an input image is constructed by a predicting unit for predicting a pixel value at a position which is more detailed than a pixel position by using a class sort adaptive process to a target image signal; and a motion vector detecting unit for detecting the motion vector for the image signal including the predicted pixel value.

Abstract: A method and apparatus for assigning a temporary label to each connected area in an image by scanning the image by using a window which has a size of two pixels in the vertical direction and of a plurality of pixels in the horizontal direction. A set of values of pixels contained in the above window is obtained and one of predetermined temporary label assignment rules corresponding to the obtained set of pixel values is selected. A temporary label is assigned to each pixel contained in the window, based on the above one of the temporary label assignment rules determined as above, and based on temporary labels of pixels in the second group in the window at the above each location. In addition, the temporary labels are converted to true labels, by scanning the image pixel within the at least one circumscribing area only, where each circumscribing area is predetermined so that the at least one circumscribing area contains all pixels which do not belong to a background area in the image.

Abstract: A method codes a hierarchical subband system of coefficients of a wavelet transform that is representative of an image. One step of this method includes receiving the coefficients of a wavelet transform. A second step includes selecting a first plurality of thresholds ranges. A third step includes comparing the coefficients of a tree structure of coefficients against the plurality of threshold ranges during a first single traversal of the tree structure. A fourth step includes coding the coefficients of the tree structure with symbols based on whether the coefficients are or are not within at least one of the plurality of threshold ranges. A method may include defining a set of symbols for coding a hierarchical subband system of coefficients of a wavelet transform that is representative of an image.

Abstract: This is directed to a method and a system for efficiently performing an image input positional adjustment prior to image or character recognition for data entry into a computer system.

Abstract: An image is divided into blocks of N consecutive pixels, and each block is coded with reference to M reference pixels, where M and N are integers greater than one. Different patterns of N pixels are obtained from the M reference pixels. If one of these patterns matches the N consecutive pixels sufficiently closely, that block of N consecutive pixels is coded as a flag and a pattern code. When the coded data are decoded, the flags identify blocks that were coded in this way, and these blocks are decoded by copying already-decoded data identified by the pattern codes.