Showing papers on "Image gradient published in 1994"

Representing moving images with layers

Abstract: We describe a system for representing moving images with sets of overlapping layers. Each layer contains an intensity map that defines the additive values of each pixel, along with an alpha map that serves as a mask indicating the transparency. The layers are ordered in depth and they occlude each other in accord with the rules of compositing. Velocity maps define how the layers are to be warped over time. The layered representation is more flexible than standard image transforms and can capture many important properties of natural image sequences. We describe some methods for decomposing image sequences into layers using motion analysis, and we discuss how the representation may be used for image coding and other applications. >

Image processing apparatus using pattern generating circuits to process a color image

Abstract: In the apparatus for processing a color image, a data signal representing another information differs from the color image is generated by a generating means. The another information is embedded into the color image by varying one of a color difference and a chroma of the color image in accordance with the data signal by an image processing means.

Contrast enhancement of medical images using multiscale edge representation

Abstract: Experience suggests the existence of a connection between the contrast of a gray-scale image and the gradient magnitude of intensity edges in the neighborhood where the contrast is measured. This observation motivates the development of edge-based contrast enhancement techniques. We present a simple and effective method for image contrast enhancement based on the multiscale edge representation of images. The contrast of an image can be enhanced simply by stretching or upscaling the multiscale gradient maxima of the image. This method offers flexibility to selectively enhance features of different sizes and ability to control noise magnification. We present some experimental results from enhancing medical images and discuss the advantages of this wavelet approach over other edge-based techniques.

Image processing system suitable for colored character recognition

Abstract: An image processing system capable of facilitating the highly accurate character recognition on the colored input images. In this system, an input image data to be processed is entered, and a color separation of the input image data is determined according to geometrical data for each image portion of the input image data, and color data for each image portion and a background of the input image data. Then, the input image data is appropriately processed according to the determined color separation. Also, in this system, at least one of color image data and gray scale image data according to the input image data are stored along with and binary image data according to the input image data, and the binary image data are processed by looking up the at least one of the color image data and the gray scale image data.

Image simulation method

Abstract: Simultaneously and automatically with creating a CG image by computer graphics, (1) the image is segmented to create a mask image, (2) a mask attribute table is created by data transformation, (3) an intrinsic image (object color component, light source color component, ambient color component) is created by data transformation, and the results are saved. The mask image is used to cut a portion from the computer graphics image and the cut portion is combined with a different image, which may be a natural image for example, to obtain a composite image, and at the same time, transparency processing or other such image processing is conducted on the basis of the attribute data. In addition, the color etc. of the object are modified by changing the object color vector or the like.

Functional interpolating transformation system for image processing

Abstract: An image processing system and a concomitant method for rapid image processing. The system generates special image format containing a plurality of subimages. When an image is edited, the system uses the information in the subimages to calculate a depiction of the image on the display appropriate to display resolution, given the current pan and zoom. Image alterations are stored within a computer file as an expression tree. Each node in the expression tree defines a particular editing operation that resulted in modifications to the image. To display the image upon a computer monitor display, the expression tree defining the alterations is evaluated to form a composite output image having a resolution appropriate to display an accurate rendition of the edited image. Once edited, the output image can be printed on a high resolution printer by combining the expression tree with a full resolution image to produce a high resolution output image.

Image signal processing device

Abstract: An image signal processing device generates first image data and second image data based on the same optical image. Each pixel of the second image data is offset from the corresponding pixel of the first image data by half the distance between the centers of two adjacent pixels. The first image data is then subjected to a discrete cosine transformation (DCT), quantization, and Huffman encoding, and is recorded to an IC memory card. High resolution image data is generated based on the first and second image data. Expanded image data is obtained based on the first image data, and supplementary data is generated based on the high resolution image data and the expanded image data. The supplementary data is subjected to DCT, quantization and Huffman encoding, and is recorded to the same IC memory card.

Image processor with edge emphasis of image data

Abstract: Image data of an original image in a prescribed pixel matrix is subjected to edge emphasis. On the other hand, a maximum and a minimum of the image data are detected. Then, values of the image data subjected to edge emphasis are limited between the maximum and the minimum to provide output data. Thus, edge emphasis can be performed without increasing noises even for hand-written characters of low densities.

Locating transition states by quadratic image gradient descent on potential energy surfaces

Abstract: An analysis is given of the so‐called ‘‘image function’’ approach to finding transition states. It is demonstrated that, in fact, such functions do not exist for general potential energy surfaces so that a plain minimum search is inappropriate. Nonconservative image gradient fields do exist, however, and their field lines, defined by Euler’s equation, can lead to transition states as exemplified by quantitative integrations of these equations for the Muller–Brown surface. As do gradient fields, image gradient fields contain streambeds and ridges, but their global structure is considerably more complex than that of gradient fields. In particular, they contain certain singular points where the image gradients change sign without passing through zero. They are the points where the two lowest eigenvalues of the Hessian are degenerate. Some of them can act as singular attractors for the image gradient descent and any algorithm must contain safeguards for avoiding them. (Such regions are equally troublesome for quasi‐Newton‐type transition‐state searches.) Image gradient fields appear to have considerably larger catchment basins around transition states than do quasi‐Newton‐type or gradient‐norm‐type transition‐state searches. A quantitative quadratic image‐gradient‐following algorithm is formulated and, through applications to the Muller–Brown surface, shown to be effective in finding transition states.

Histogram-based morphological edge detector

Abstract: Presents a new edge detector for automatic extraction of oceanographic (mesoscale) features present in infrared (IR) images obtained from the Advanced Very High Resolution Radiometer (AVHRR). Conventional edge detectors are very sensitive to edge fine structure, which makes it difficult to distinguish the weak gradients that are useful in this application from noise. Mathematical morphology has been used in the past to develop efficient and statistically robust edge detectors. Image analysis techniques use the histogram for operations such as thresholding and edge extraction in a local neighborhood in the image. An efficient computational framework is discussed for extraction of mesoscale features present in IR images. The technique presented in the present article, called the Histogram-Based Morphological Edge detector (HMED), extracts all the weak gradients, yet retains the edge sharpness in the image. A new morphological operation defined in the domain of the histogram of an image is also presented. An interesting experimental result was found by applying the HMED technique to oceanographic data in which certain features are known to have edge gradients of varying strength. >

Color image input apparatus having color image identifying function

Abstract: A line sensor and image input section input a color image in a reading area containing an original, an original extracting section extracts an area of the original from the input image, a normalization section normalizes an image of the extracted original area to an image of preset size, an image averaging section converts the normalized image into an averaged image, and a brightness-hue-chroma converting section converts the averaged image in Vcd images used as color perception amounts of a human being. A pattern matching section collates the Vcd image with dictionary data in a dictionary data storing section to determine whether or not the original is a specified type of original such as a bill, and an image output controlling section determines whether or not image data output from the color image input section is to be output to the exterior based on the result of identification.

Imae processing apparatus having mosaic processing feature that decreases image resolution without changing image size or the number of pixels

Abstract: A copying apparatus includes an input device for inputting image data, a processing circuit for performing mosaic processing of the input image data, and a reproduction circuit for reproducing an image based on the processed image data. The processing circuit divides the input image data into a plurality of block areas and paints each block area with a uniform color according to the image data in the area so that the resolution of the image data is lower than the resolution of the original image.

A region-region and region-edge cooperative approach of image segmentation

Abstract: In the task of segmentation of some complex pictures, it is often difficult to obtain satisfactory results using only one approach of image segmentation. The tendency toward the integration of several techniques seems to be the best solution. The authors introduce a new approach of image segmentation, which tends to combine several sources of knowledge about the image in a way to produce better segmentation results. First, they try to locate germs that are homogenous by means of a region-region cooperative process. A region growing process is then applied to these germs in order to find the region borders. This process is controlled, on the one hand, by the germs' parameters, and on the other hand by a gradient information obtained by a simple edge detector. Finally, a region merging step is applied on the extracted regions in order to reconstruct regions that have been split by the germs' extraction process. This method has given good segmentation results over several complex natural images. >

Edge detection using the local fractal dimension

Abstract: Fractal Brownian noise is used as a model describing the local grey level change in digital images. At edges this model does not truly reflect the reality, because edges add a deterministic component to the image which is not compatible with the notion of scale-independent self-similarity of fractal structures. Thus, the local degree of 'fractality' is used to differentiate edges from segment interiors and from noise. The concept is evaluated by comparing fractal edge detectors with conventional operators such as, e.g., a Sobel or Laplace operator. Results show a similar performance in a low-noise environment and superiority of the fractal operators in a high noise environment. The inclusion of the operators into an edge-based segmentation scheme revealed the same results for an application in image segmentation. >

An image database system with fast image indexing capability based on color histograms

Abstract: While general object recognition is difficult, it is relatively easy to capture some inherent image properties, such as color distribution, to narrow down the search range when an attempt to retrieve images from an image database is made. We have built an image database in which images are indexed automatically using a histogram indexing method that we have proposed. With this method, a numerical index key is created for each image histogram using a set of mathematical formulas. By using this approach, we have transformed the difficult problem of image matching into one of image retrieval by index keys, which can be supported by existing database implementation techniques. We have also provided two query methods: image retrieval by user-provided sample images, and image retrieval by combinations of system-provided templates (such as sky, lawn, brick wall, and so on). Our initial experiments on the image database system have shown favorable results. >

Multichannel edge enhancement in color image processing

Abstract: In the paper, a multichannel edge enhancing filter (MEEF) based on the vector median is introduced for enhancing degraded edges in color images. An input multichannel signal is filtered with three subfilters, and the final output is determined by comparing the outputs of the subfilters and their vector median. Root signal and edge enhancement properties of the MEEF are examined in detail. In addition, the authors discuss line preservation and edge jitter resistance of the MEEF in noisy conditions. Finally, the MEEF is tested with: i) blurred edges caused by unfocusing or fast movement of camera, ii) serrated edges caused by interlaced scan, iii) false color edges caused by channel dispersion. The results show that the MEEF outperforms channelwise edge enhancing filtering in enhancing these degraded edges. >

Method and apparatus for improving motion analysis of fades

Abstract: A method and apparatus for improving motion analysis in fade regions where motion compensation systems based on block matching typically fail to accurately encode images. A previous image is selected from an image sequence. A base image is calculated by adjusting the brightness of the previous image. A displacement vector representing the magnitude and direction of the displacement between at least one region in a target image and a corresponding region in the base image is determined. The displacement vector is applied to a corresponding region in a previous reconstructed image to form a predicted image, and is then encoded. Pixel values in the predicted image are subtracted from corresponding pixel values in the target image to form an error image which is also encoded.

Better geometric measurements based on photometric information

Abstract: We propose new methods for estimating properties of analog objects in properly sampled multi-dimensional grey-scale images. The finite aperture of lenses ensures band limitation of the analog image and allows sampling. Many existing measurement procedures work on a binary object obtained by edge detection and thresholding. The ragged binary edge is disturbed by aliasing which cannot be repaired by smoothing. To solve this problem we propose methods that work directly on the grey-scale image. The grey-scale image contains accurate photometric information. Our new methods yield errors that are generally an order of magnitude better than the traditional binary ones. For applications where a smooth, constant edge height is a prerequisite we introduce erf-clipping. Erf-clipping is a point operation that shapes a linear edge region into a scaled error function. In contrast to thresholding it requires very mild oversampling. >

Normalization method for eliminating false detections in side scan sonar images

Abstract: A method for eliminating false detections at the water column boundary in side scan sonar images is provided. Each image is filtered by computing the median greylevel value of a pixel and its four adjacent pixels to produce a median image. The median image is filtered with a first linear filter to produce an edge image. The edge image is then filtered with a second linear filter to produce an edge-smoothed image. The edge of the image is detected by searching each row of the edge-smoothed image for the maximum greylevel. A transition region is defined between the detected edge and a preselected right boundary and the average greylevels over the rows and columns in this transition region is computed. The sonar image is then normalized by computing the image mean grayness level of each pixel in the image.

Method and apparatus for identifying the color of an image

Abstract: A method and an apparatus to reduce the errors in linearly transforming the RGB values to the XYZ values of the colors of an image. The image is measured twice, once with and once without a filter between the image and the sensors measuring the image. The apparatus is capable of generating at least three output values substantially simultaneously to measure the color of each element of the image.

Image enhancement using the log-ratio approach

Abstract: In this paper, we present a new image enhancement algorithm for interactively modifying the contrast and the sharpness of an image. The proposed algorithm aims at solving two basic problems associated with typical image enhancement technique: the enhanced image is usually out of the gray level range of an image, and there are conflicting requirements for image sharpening and noise reduction. A vector space based approach, the log-ratio approach, is employed to solve the first problem and a multiscale approach is used for the second problem. The proposed algorithm has been used successfully to enhance medical images. >

Picture processor for vehicle

Abstract: PURPOSE: To facilitate an operation, to speed up a processing and to reduce the constitution scale of hardware by permitting an edge detection means to shift an image obtained by means of an image pickup means in horizontal/ vertical directions by the prescribed number of image elements, to compare the luminance of the image before and after shifting and to detect edge information in an image state CONSTITUTION: A vehicle image processor is provided with a CCD camera being the image pickup means which image-picks up the periphery of a vehicle, an image memory storing the image outputted from the CCD camera and CPU executing the image processing based on data of the image memory At the time of processing the image, the inputted image 31 is shifted in the horizontal/ vertical directions by n-pixels and m-pixels The image signal of the shifted image 32 is compared with that of the image 31 before shifting A threshold Vth34 is set for the compared result 33 of the images 31 and 32 Binarization is executed based on the threshold and a signal 35 is obtained as edge information COPYRIGHT: (C)1995,JPO

Method and apparatus for accurately rendering half-bitted image pixels

Abstract: A method and apparatus for identifying regions within a first binary image where half-bitting may be present, converting those regions to a multiple-bit/pixel representation so as to accurately represent the image density and the intended edge structure for the region, and further generating an enhanced resolution representation of the region in either a second binary image, wherein the second binary image has a spatial resolution greater than the first image, or a multiple-bit per pixel (gray) image at the same resolution as the input image so as to enable an improved rendering of the first binary image.

Edge compensated transform coding

Abstract: Transform based image compression has difficulty with image regions containing edges. Edge compensated transform coding (ECTC) addresses this problem by preprocessing to remove edges. This preprocessing is adapted to transform coding. The edge information is sent in a side channel and the edges are replaced at the receiver. Subjective improvement is demonstrated. >

A step edge detector algorithm based on symbolic analysis

Abstract: This paper analyzes the noise error, in both the module and the argument in the calculation of the gradient vector of the images illumination function. The results show that the behavior of the argument is more robust than that of the module. A proposal is made for a symbolic analysis of the argument of the gradient vector to detect the contour of the objects in the image. This shows that it is possible to use smaller windows to calculate the gradient vector without infringing the contradiction proposed by Marr and Hildreth (1980) and Canny (1983). Finally, a description is given of an edge detection algorithm where their most important characteristics are: a) it introduces a symbolic analysis of the argument of the gradient vector to detect edges; and b) It uses smaller window to approximate the value of the gradient vector allowing to locate the edge with precision.

Image recognition method

Abstract: An image recognition method is conducted by extracting characteristic points of the pattern of an image formed by video signals from an image pickup device, within the whole or a limited area of a frame of a display, and measuring the distribution of the characteristic points of the image pattern. With this method, images can be recognized with minimal image processing time and with reduced influence of noises.

A novel edge detector based on nonlinear local operations

Abstract: A new type of edge detector is presented that uses spatial filtering of the locally thresholded image. The process is realized in two main stages. In the first one an appropriate threshold is used and the multilevel image is transformed to binary one. The second stage verifies the existence of an edge point by correlating the binary valued image points with certain binary valued spatial patterns. False edge points generated by noise are removed by means of another complementary filtering path based on the edge step size. >