166 results about "Pixel value difference" patented technology

Numerical image processing of two or more medical images to provide grayscale registration thereof is described, the numerical image processing algorithms being based at least in part on a model of medical image acquisition. The grayscale registered temporal images may then be displayed for visual comparison by a clinician and/or further processed by a computer-aided diagnosis (CAD) system for detection of medical abnormalities therein. A parametric method includes spatially registering two images and performing gray scale registration of the images. A parametric transform model, e.g., analog to analog, digital to digital, analog to digital, or digital to analog model, is selected based on the image acquisition method(s) of the images, i.e., digital or analog/film. Gray scale registration involves generating a joint pixel value histogram from the two images, statistically fitting parameters of the transform model to the joint histogram, generating a lookup table, and using the lookup table to transform and register pixel values of one image to the pixel values of the other image. The models take into account the most relevant image acquisition parameters that influence pixel value differences between images, e.g., tissue compression, incident radiation intensity, exposure time, film and digitizer characteristic curves for analog image, and digital detector response for digital image. The method facilitates temporal comparisons of medical images such as mammograms and/or comparisons of analog with digital images.

Provided are an apparatus and method for generating panorama images and an apparatus and method for tracking an object using the same. The apparatus for generating panorama images estimates regional motion vectors with respect to respective lower regions set in an image frame, determines a frame motion vector by using the regional motion vectors based on a codebook storing sets of normalized regional motion vectors corresponding to a camera's motion and then accumulates motion mediating variables computed with respect to respective continuous image frames to determine an overlapping region, and matches the overlapping regions to generate a panorama image. The apparatus for tracking an object sets the panorama image generated by the apparatus for generating panorama images as a background image, and determines an object region by averaging pixel value difference values between pixels that constitute a captured input image and corresponding pixels in the panorama image. Regional motion vectors are estimated by using the coordinate of a pixel selected in lower regions, and thus a computing process can be simplified and a process speed can be increased. In addition, effects of wrongly estimated regional motion vectors can be minimized by using a codebook. Furthermore, noises are minimized and an object can be accurately extracted by using averaged pixel value difference values.

After a plurality of differently exposed images are obtained, and tone levels of the plurality of images are adjusted, image elements corresponding to changes of objects are detected based on pixel value differences between the plurality of images. Then, the numbers of blown-out highlight and shadow-detail loss image elements in the image elements corresponding to the changes of the objects are counted for the plurality of images. Then, an image in which the total of the numbers of image elements is smallest is selected. Furthermore, image elements corresponding to the changes of the objects in the selected image are used as those corresponding to image elements corresponding to the changes of the objects in an HDR image to be generated.

There are provided an apparatus and method for performing dynamic capacitance compensation (DCC) in a liquid crystal display (LCD). The DCC apparatus includes: a first line buffer reading and temporarily storing pixel values of an image for each line; an encoder transforming and quantizing the pixel values stored for each line for each block and generating bit streams; a memory storing the generated bit streams; a decoder decoding the bit streams stored in the memory for the each block and outputting the decoded bit streams; a second line buffer reading and temporarily storing the decoded pixel values for the each block; and a compensation pixel value detector detecting a compensation pixel value for each pixel, from pixel value differences between pixel values of a current frame stored in the first line buffer and pixel values of a previous frame stored in the second line buffer. Therefore, it is possible to reduce the number of pins of a memory interface by reducing the number of memory device for storing pixel values of image data, required for performing DCC of a LCD, resulting in minimizing a chip size, and to enhance compression efficiency without visual deterioration in images.

A method for reducing blocking artifacts in a video stream comprises receiving a picture of the video stream, wherein the picture includes a plurality of macroblocks and each of the plurality of macroblock includes four blocks, determining blocks with quantization parameters greater than a first threshold value in the picture, checking if block boundaries of the blocks are sharp and are real edges of objects in the picture according to pixel value differences between two adjacent pixels respectively located at both sides of the block boundaries, selecting filtering strengths of a de-blocking operation according to the pixel value differences when the block boundaries are sharp and are not real edges of the objects in the picture, and performing the de-blocking operation for two adjacent blocks at both sides of the block boundaries.

A fast loss less image compression system based on neighborhood comparisons compares pixel value differences with neighboring pixels and replaces such pixel values with the minimum of the differences. A marker is attached to a block of pixels, such that all the pixels in that block are compared with neighbors of one direction. The marker indicates how all of the pixels in that block are compared. Intermittent Huffman-tree construction is used such that one tree is used for several frames. Huffman coding is used to compress the resulting frame. A single Huffman-tree is constructed once every predetermined number of frames. The frequency of Huffman-tree construction can be performed according to the instantaneous availability of processor time to perform the construction. When more processing time is available, the Huffman-trees are computed more frequently. Such frequency variation can be implemented by using an input video frame buffer. If the buffer is a certain size, then processor time for Huffman-tree construction is available.

An image processing apparatus includes an adaptive interpolation device which converts a MFA pattern image into a quincuncial pattern image based on difference values, and interpolates color channels and an NIR channel, based on difference values of the converted quincuncial pattern image in vertical and horizontal pixel directions; a frequency compensation device which obtains a high-resolution MFA image using high-frequency and medium-frequency components of a high-resolution base image, based on linear regression analysis and compared energy levels of MFA channel images to an energy level of a base image; and a channel interference suppression device which removes color distortion generated between each channel of the high-resolution MFA image, and another channel of the high-resolution MFA image and a base channel using a weighted average of pixel value differences between each channel of the high-resolution MFA image, and the other channel of the high-resolution MFA image and the base channel.

An image processing apparatus, which correctly extracts specular reflected light components of reflected light from an object, and accurately estimates a light source color, is provided. The image processing apparatus calculates a pixel value difference distribution by repeating, for respective pixels, to calculate pixel value differences between a pixel of interest and adjacent pixels in an input image and to calculate similarities between pixel value differences. A light source estimation unit estimates a color of a light source which illuminates an object in the input image based on the calculated distribution.

The invention provides a moving target detection method and device, and electronic equipment. The method comprises the following steps: determining a current frame image and a specified frame image corresponding to the current frame image from a to-be-detected video according to the sequence of each video frame in the to-be-detected video; determining a detection area of the moving target according to the pixel value difference of the corresponding pixel points between the current frame image and the specified frame image; inputting the image data corresponding to the detection area into a preset target detection model, and outputting an initial detection result corresponding to the detection area; and mapping the initial detection result to the current frame image to obtain a final detection result. According to the embodiment of the invention, the rough detection area of the moving target can be obtained through the pixel value difference between the two frames of images, thereby reducing the detection range of the target detection model, and improving the detection efficiency. Meanwhile, most static targets are eliminated from the detection area, so that the interference on dynamic target detection of the model is reduced, and the detection accuracy is improved.

The invention discloses a parking event detecting method based on a double tracking system. By means of segmentation of a plurality of block areas and drawing of a dynamic background, the sum of absolute values of pixel value differences of each area of the background corresponding to a current frame is calculated, the steps are repeated, a suspicious area is marked and assigned a value 255, a connected domain analysis is conducted, and whether a car is parked on the suspicious area is confirmed. The parking event detecting method based on the double tracking system is free from environmental constraints, a real-time detection can be conducted, the impacts of shadows, lights, threw objects and surface gathered water can be eliminated preferably by means of the combination of the analysis of tracing lines, the algorithm can achieve real-time and accurate detection of road unusual parking events, rescue and handling of a traffic accident can be conducted timely and effectively, a secondary accident is prevented from happening, and the safety of road operation is further guaranteed.

The present invention relates to a dynamic face identification system and method. The system comprises a detection module, a pre-processing module, a feature extraction training module, a feature extraction identification module. The detection module loads a face detector, reads a video stream or a panoramic image to detect the faces and displays, intercepts and saves the detected face images real-timely, the pre-processing module carries out the grey level transformation and the image normalization on the face images, detects by the accumulation of pixel values difference in wavelet domain (APVD) and retains the face images reaching a standard, the feature extraction training module establishes a training sample base and the indexes of the training samples, reads the face images reaching the standard to extract the features, and carries out the PCA feature dimensionality reduction and the BP neural network training, and the feature extraction identification module reads the face images reaching the standard, carries out the normalization, the curve Gabor wavelet (CGW) feature extraction and the PCA dimensionality reduction and identification, and outputs an identification result. According to the present invention, by calculating the fuzzy degree of the images, selecting the face images reaching the identification requirements, and using a curve Gabor wavelet to extract the effective face features, the identification rate is improved.

A vision based pedestrian and cyclist detection method includes receiving an input image, calculating a pixel value difference between each pixel and the neighbor pixels thereof, quantifying the pixel value difference as a weight of pixel, proceeding statistics for the pixel value differences and the weights, determining intersections of the statistics as a feature of the input image, classifying the feature into human feature and non-human feature, confirming the human feature belonging to cyclist according to the spatial relationship between the human feature and the detected two-wheeled vehicle, and retaining one detection result for each cyclist by suppressing other weaker spatial relationships between the human feature and the detected two-wheeled vehicle.

The present invention relates to the fields of motion detection, scene change detection and temporal domain noise reduction. Especially, the present invention relates to a device, a method and a software for detecting motion and/or scene change in a sequence of image frames. The device for detecting motion and/or scene change in a sequence of image frames comprises a difference generator adapted to calculate one or more differences between values of pixels of separate image frames, the pixel values comprising color information, and a discriminator for providing an indication whether the calculated one or more pixel value differences correspond to noise or correspond to image content based on the one or more pixel value differences.

A method of generating an intermediate image between a first frame and a second frame, using the first frame and a third frame successively received from a first camera and the second frame and a fourth frame successively received from a second camera. The intermediate image generating method includes: calculating and summing pixel value differences between the first frame and the third frame in units of a block; determining whether or not a center pixel of a block has motion; calculating current disparity vectors corresponding to motion vectors between the first frame and the second frame, and calculating previous disparity vectors corresponding to motion vectors between the third frame and the fourth frame; and generating the intermediate image using the current disparity vectors if the center pixel of the block has motion, and calculating an intermediate value of the previous disparity vectors and generating the intermediate image using the intermediate value of the previous disparity vectors if the center pixel of the block has no motion.

An apparatus and method are provided which produce a precise evaluation value table to realize an exact motion vector detection process. A weighting factor (W) is calculated based on information of correlation between representative point pixels and on flag correlation information based on flag data corresponding to pixel value difference data between a pixel of interest and its neighboring area pixels. A reliability index is produced based on the calculated weighting factor (W) and on an activity (A) serving as an indicator value indicative of complexity of picture data. Alternatively, a reliability index is produced based on a motion similarity between a representative point pixel and its neighboring pixels. An evaluation value table is produced in which evaluation values corresponding to the reliability indices have been integrated. In addition, a correlation determination is executed based on the pixel values or positions of featuring pixels on the periphery of the pixel of interest to determine a motion vector.

A method of detecting camera tempering and a system therefor are provided. The method includes: performing at least one of following operations: (i) detecting a size of a foreground in an image, and determining whether a first condition, that the size exceeds a first reference value, is satisfied, (ii) detecting change of a sum of the largest pixel value differences among pixel value differences between adjacent pixels in selected horizontal lines of the image, according to time, and determining whether a second condition, that the change lasts for a predetermined time period, is satisfied, and (iii) adding up a plurality of global motion vectors with respect to a plurality of images, and determining whether a third condition, that a sum of the global motion vectors exceeds a second reference value, is satisfied; and determining occurrence of camera tempering if at least one of the corresponding conditions is satisfied.

The invention discloses a method for generating depth maps of images. The method includes 1), selecting a plurality of reference blocks in corresponding regions of a reference-frame image for a current block of a current-frame image; 2), computing a depth block matching error and a color block matching error between the current block and each reference block; 3), weighting the corresponding depth block matching error and the corresponding color block matching error for the current block and a certain reference block to obtain a comprehensive matching error; and 4), selecting a certain reference block corresponding to the minimum comprehensive matching error and using a depth map of the reference block as a depth map of the current block. Each color block matching error is the sum of absolute values of color pixel value differences among all corresponding pixels between the current block and the corresponding reference block. The method has the advantages that the quality of the depth map obtained after the current-frame image is interpolated is improved, and accumulative errors caused by one-way depth interpolation propagation are effectively reduced.

A method and apparatus to automatically control power of address data in a PDP includes a plurality of address electrodes, a plurality of scan electrodes, and a plurality of sustain electrodes arranged in pairs with the scan electrodes, and a PDP including the apparatus. The sum of pixel value differences between adjacent ones of successive lines in input image data is calculated, and an address power control (APC) level corresponding to the calculated line pixel value difference sum is determined. The image data is then repeatedly multiplied by a gain initially corresponding to the start gain and sequentially decremented by a predetermined value from the start gain upon every multiplication until the gain corresponds to the end gain, to output corrected address data.

An embodiment of the invention discloses a training model generation method comprising the following steps: a plurality of human face positive samples and human face negative samples are obtained, characteristics of pixel difference between each pixel in each sample and other pixels of the sample are calculated, one object pixel difference characteristic is chosen from all the calculated pixel difference characteristics, point coordinates and pixel value difference corresponding to the object pixel difference characteristic are set as decision nodes for a decision making tree and are used for determining and distinguishing the positive samples and the negative samples, the decision nodes of the decision making tree are used as weak classifiers that are then cascaded, steps from a step of choosing the object pixel difference characteristic from all the calculated pixel difference characteristics to a step of using the decision nodes of the decision making tree as the weak classifiers that are then cascaded are subjected to iterative execution operation, and finally a strong classifier is generated. Point coordinates and pixel value difference corresponding to each decision node of the decision tree in the strong classifier are stored, and therefore a training model can be generated. The strong classifier is formed via the decision tree, a human face can be subjected to secondary classifying operation, and time needed for human face detection can be reduced when the same precision remains unchanged.