889 results about "Edge maps" patented technology

A hybrid compression scheme in which one or more compression schemes are applied to a scanned document depending on its content. Based on the scanned document data itself and edge-map version thereof, a color test is performed to determine if the document is text or a drawing. If only a few colors are detected, the entire document is compressed by a photo compression technique at full resolution. If more than a few colors are present, the document is most likely a mixed mode document, in which case the entire document undergoes segmentation, a process in which text and drawing regions are separated from the background. The mask of the text and / or drawing regions are coded by a bi-level compression technique, whereas background is down-sampled, if necessary, and then compressed by a photo compression technique. A wrapper module is used to wrap the compressed stream(s) into a single file for easy browsing or transmitting.

A system and method for processing a digital video signal corresponding to an image are provided. A plurality of independent edge detecting processes or edge detector modules detect a set of edges and at least one additional edge that is not included in the set of edges. An edge map includes data regarding all edges identified by any edge detecting process or module, and a visually perceptible artifact of the image is altered based at least in part on an evaluation of the edge map. The system and method detects and filters block artifacts and ringing or other noise from digital images, resulting in reduced image distortion.

An image stabilizing device has an image restoration processor configured to perform image restoration process for stabilizing hand movement included in an input image by using an image restoration filter created based on hand movement information, and an edge emphasizing processor configured to perform edge emphasizing process for emphasizing an edge of the input image. The image stabilizing device has a stabilization mode determiner configured to determine an image stabilization mode for stabilizing the input image based on a size of the hand movement created by the hand movement information. The image stabilization mode includes a first mode for performing the image restoration process, a second mode for performing the edge emphasizing process, and a third mode for performing none of the image restoration process and the edge emphasizing process.

A motion vector calculation unit 4 calculates inter-frame movement amounts. A masked region specification unit 5 separates the entire edge image of frame feature data into (i) an edge image showing relatively large movement amounts and (ii) an edge image showing relatively small movement amounts. The masked region specification unit 5 then specifies the edge image showing relatively large movement amounts as a region to be masked. This way, a correction parameter is detected from a region other than the masked region. When the correction parameter is a slang angle, a slant correction unit 7 performs slant correction on a frame picture obtained by an image sensor.

The invention relates to a tunnel lining crack width measuring method, which comprises the follow steps of: (1) picking up crack images through a digital camera and measuring pickup distance through a laser distance measuring instrument; (2) selecting a region required to be measured on the picked images and converting selected region images into gray level images; (3) determining the threshold value of the gray level images selected in the second step, cutting binary image cutting, and obtaining a target region; (4) extracting a white region only comprising crack in the target region; (5) extracting the edge of the crack by a sub pixel method, and obtaining sub pixel level edge images; (6) circulating the pixel width of the crack by a minimum distance method; (7) circulating the actual width of the obtained crack according to the calibration proportion of the predetermined actual pixel size and the image pickup distance; and (8) storing the obtained actual width data into a crack data base. Compared with the prior art, the tunnel lining crack width measuring method has the advantages of low cost, high speed, convenience, high precision and the like.

The invention discloses a target detection and recognition method for water surface panoramic images, and belongs to the technical field of computer visual recognition. The method comprises the following steps of: firstly carrying out target detection and recognition on a panoramic image by utilizing a target detection and recognition neural network model, and obtaining a target type and an initial position of a detection box; carrying image segmentation on a target local area to obtain a plurality of superpixel blocks, and combining the superpixel blocks by adoption of a multi-threshold fusion strategy so as to obtain an adjusted target detection box; calculating an edge map of a new target local area, adjusting the target detection box according to peripheral edge points of the target detection box so as to obtain a final target detection box; and finally converting the position of the final target detection box into a target practical position. According to the method, distortion effect in panoramic images can be effectively overcome, and target positions can be correctly recognized from the panoramic images.

A image processing system for multi-label semantic edge detection in an image includes an image interface to receive an image of a scene including at least one object, a memory to store a neural network trained for performing a multi-label edge classification of input images assigning each pixel of edges of objects in the input images into one or multiple semantic classes, a processor to transform the image into a multi-label edge-map using the neural network detecting an edge of the object in the image and assigning multiple semantic labels to at least some pixels forming the edge, and an output interface to render the multi-label edge-map.

A method for correcting colored fringe artifacts in a region of an image around saturated pixels captured by an electronic imager associated with a color filter array, includes providing a threshold map of the image based on thresholding one or more color channels with a common or separate threshold value for each color channel; providing a first edge map by dilating the threshold map as a function of the color fringe width, producing a first dilated threshold map and subtracting the threshold map from the first dilated threshold map; and color-desaturating image pixels that correspond to the first edge map.

A mobile device uses an image-driven view management approach for annotating images in real-time. An image-based layout process used by the mobile device computes a saliency map and generates an edge map from a frame of a video stream. The saliency map may be further processed by applying thresholds to reduce the number of saliency levels. The saliency map and edge map are used together to determine a layout position of labels to be rendered over the video stream. The labels are displayed in the layout position until a change of orientation of the camera that exceeds a threshold is detected. Additionally, the representation of the label may be adjusted, e.g., based on a plurality of pixels bounded by an area that is coincident with a layout position for a label in the video frame.

A method of compressing video content is disclosed and may include receiving an image frame of a video signal having multiple components, generating an edge-map for the image frame, generating a bitstream base layer for each component of the video signal, generating a first bitstream enhancement layer for each component of the video signal, and generating successive bitstream enhancement layers for each component of the video signal. As the successive bitstream enhancement layers are generated, the method of compressing video content goes from lossy to lossless.

A conversion method and apparatus with a generating of a depth map for two dimensional (2D)-to-three dimensional (3D) conversion. A depth order may be restored based on a line tracing and an edge map generated from an input image, and a stereo image may be generated using depth information.

The invention discloses an adaptive threshold edge detection algorithm based on canny. The adaptive threshold edge detection algorithm based on canny includes that firstly, carrying out Gaussian smoothing on an image, calculating the corresponding gradient, and restraining the maximum value of a gradient map; calculating to obtain a statistical histogram of the gradient map according to the detected gradient size, expressing gradient background according to the histogram information, and calculating to obtain the high threshold and low threshold of the gradient map to obtain a strong edge map and a weak edge map; using the edge point of the strong edge map as a central point to compare the edge information of the field edge point and the corresponding position in the weak edge map to connect strong and weak edges. Compared with a traditional method which sets the proportion of the strong edge as a fixed value and a method which obtains the high threshold through an inter-class maximum-variance algorithm and obtains the low threshold through multiplying the high threshold by a fixed proportion, based on keeping the low complexity and high edge quality of the original canny method, the adaptive threshold edge detection algorithm based on canny further improves the edge continuity and precision, is more suitable for extracting the main outline, and provides basis for the subsequent high-precision target detection and target division.

Aspects of depth map generation and post capture focusing and re-focusing are described. According to one embodiment, a depth map is generated. The depth map may include a mapping among relative depth values in a field of view of an image based on a difference between pixels of a first image and pixels of a second image. An edge map may also be generated by identifying edges in at least one of the first image or the second image. Using the depth map and the edge map, the relative depth values in the depth map may be smoothed using the edge map. In this manner, certain discontinuities in depth values may be smoothed within edge-bounded regions defined by the edge map. The depth map may be used for focusing and re-focusing, for example, or for object extraction, scene understanding, or gesture recognition, among other imaging processes.

The invention discloses a color image segmentation method, which comprises the following steps: processing an image to be segmented into a gray image; extracting a region profile map from the gray image by an Otsu threshold segmentation method, and determining an optimal threshold; taking the optimal threshold as a high threshold of a Canny operator, and extracting an edge map from the gray image by utilizing the Canny operator; and fusing the area profile map and the edge map, and outputting a segmentation result of a color image. The method adopts an operator threshold determining scheme combined with an adaptive strategy and an empirical value and overcomes the defect that an empirical value selecting method has application limitation; the local change intensity of a gray value is obtained through edge detection, so that over-segmentation of avoidable areas is limited; and residual edges are complemented through region segmentation, so that profiles of images after the segmentation are more clear and complete.

The invention discloses a multiple lane line detecting method based on isometric peripheral point matching. The method comprises the following steps: collecting an image of a road in front of an automobile, conducting inverse perspective projection change on the collected image of the road in front of the automobile by utilizing the interrelation of world coordinate, camera coordinate and image coordinate to obtain an aerial view of the image of the road, conducting edge detection on the aerial view, conducting pretreatment on the detected edge view, conducting local area matching on all edge points of the pretreated edge view, reserving successfully matched edge point pairs, obtaining a feature point diagram through the edge point pairs, conducting partitioning on diagram according to distribution of the feature points, extracting all feature points of one lane line I, isolating feature point areas which belong to other lane lines II, and taking the feature point areas as the candidate area of the lane line I. According to the method, the technical problems of multiple lane line detection, lane line positioning and bending, and solid lane line and dotted lane line discrimination in the conventional method can be effectively solved.

A method for locating license plates in Guangdong and Hong Kong, the steps are as follows: capture sudden video images through a video camera system; initial positioning of license plates in mainland China and Hong Kong; detectable edges through edge detection, segment the edge image to obtain a binary image, and then After filtering, extract the initial positioning result image from the original grayscale image; Continental license plate positioning: Segment the initial positioning image with an adaptive threshold segmentation method to obtain a binary image, and then use the continuous black and white line segment jump of the Continental license plate feature, further narrow down the candidate area, and then use the aspect ratio feature of the mainland license plate to perform precise positioning; Hong Kong license plate positioning: use the adaptive threshold segmentation method to segment the initial positioning image, and then use the relevant features of the target circumscribed rectangle to accurately position; output positioning result. The positioning accuracy of the present invention is high, and can meet the on-site inspection requirements of the customs.

The invention discloses a method for realizing the positioning fitting of cover plate glass and a liquid crystal screen. The method comprises the following steps of: obtaining the image of the cover plate glass and the image of the liquid crystal screen, detecting characteristic points in the images, selecting one group of opposite angle marks through condition restriction, carrying out connection, carrying out refining treatment on the pixel characteristic values of each obtained characteristic point to obtain a sub pixel level coordinate value, obtaining a control parameter, and controlling the cover plate glass and the liquid crystal screen to move to finish monocular preliminary checkout; obtaining the images of one group of angular points of the cover plate glass and the images of one group of angular points of the liquid crystal screen, and detecting the edges of the images to obtain the image which is only provided with an edge line; finally, extracting straight lines from the edge image, screening lines to determine two edge straight lines of each required angular point, calculating the coordinate of each angular point, obtaining the midpoint coordinate and the slope of one diagonal line of the cover plate glass, the coordinate difference value of two midpoints and the inclined angle of two straight lines, and the midpoint coordinate and the slope of one diagonal line of the liquid crystal screen, the coordinate difference value of two midpoints and the inclined angle of two straight lines; and controlling the cover plate glass and the liquid crystal screen to move to enable the two straight lines to be in parallel and enable the midpoints to be subjected to superposition so as to finish the fitting of the cover plate glass of the touch screen and the liquid crystal screen.

The invention relates to a measuring method of a shield tunnel lining segment seam open width and an apparatus thereof. The method comprises the following steps: using a computer vision kit to demarcate cameras, collecting segment seam images through the digital cameras and determining a shooting distance through a laser range finder; converting the collected images into grayscale images, selecting an appropriate filtering parameter and using a Gauss filter to smooth the images in a selected region of interest (ROI); determining a maximum threshold and a minimum threshold after the images aresmoothed and carrying out Canny lag threshold segmentation so as to obtain a sub-pixel linear characteristic of the images and a segment seam edge; screening needed sub-pixel segment seam edge imagesfrom the sub-pixel linear characteristic; using a normal vertical minimum distance method to respectively calculate a largest width and a mean width according to the obtained two sub-pixel segment seam edge images; determining a conversion proportion between a pixel dimension and an actual dimension according to a predetermined internal parameter of the cameras and calculating so as to obtain an actual width of the segment seam. The method and the apparatus have advantages of low costs and high precision. The method can be used rapidly and conveniently.

An image resolution conversion method and apparatus based on a projection onto convex sets (POCS) method are provided. The image resolution conversion method comprises detecting an edge region and a direction of the edge region in an input low-resolution image frame in order to generate an edge map and edge direction information, generating a directional point spread function based on the edge map and the edge direction information, interpolating the input low-resolution image frame into a high-resolution image frame, generating a residual term based on the input low-resolution image frame, the high-resolution image frame, and the directional point spread function, and renewing the high-resolution image frame according to a result of comparing the residual term with a threshold.

The invention discloses a fast color image segmentation method based on combination of clustering analysis and an image pyramid. The method comprises the following steps: 1) constructing a Gauss and Laplace image pyramid according to a source image, and obtaining a low-resolution image and an edge image; 2) transforming the low-resolution image to an HSV (hue, saturation and value) space, obtaining the clustering number and an initial clustering center through histogram analysis, and performing clustering segmentation on the low-resolution image in the HSV space by use of a clustering algorithm; 3) performing up-sampling on the segmentation result obtained in the step 2), projecting to the original resolution, performing spatial filtering and removing the excessively small area to obtain the original resolution area segmentation result; and 4) integrating the edge image obtained in the step 1) and the area segmentation result obtained in the step 3) to obtain the final segmentation result. The method disclosed by the invention guarantees the segmentation quality while increasing the color image segmentation speed, and is suitable for a digital image processing system with high requirements on real-time property.

A method for processing an input image to reduce halo artifacts in an electrophotographic printing system. The input image is automatically analyzed to determine an edge map image indicating light-side edge regions that include edge pixels that are adjacent to edge transitions in the input image and are on a lighter side of the edge transition. At least one edge region dilation operation to the edge map image to expand the light-side edge regions in a direction away from the edge transitions. A corrected image is formed by modifying the input pixels of the input image corresponding to the expanded light-side edge regions to determine corrected pixels having corrected pixel values. The corrected image is then printed using the electrophotographic printing system.

A preprocessing section binarizes input image data and calculates a total black pixel ratio. A feature extracting section detects connected components contained in the binarized image data and detects circumscribing bounding boxes that circumscribe these connected components, respectively. Based on sizes of the circumscribing bounding boxes detected and numbers of black pixels contained therein, predetermined connected components are removed. A determining section generates an edge map by using the residual connected components, and performs two-dimensional fast Fourier transform thereon to generate spectral data. The determining section performs two-dimensional fast Fourier transform on template images to generate spectral data. The determining section determines, based on these pieces of spectral data, whether or not a circular shape is contained in the input image data.

A method of enhancing a full-color image of a scene includes capturing an image of the scene using a two-dimensional sensor array having both color and panchromatic pixels, forming an edge map in response to the panchromatic pixels, forming the full-color image in response to the captured color pixels, and using the edge map to enhance the full-color image.

A method reconstructs a depth image encoded as a base layer bitstream, and a set of enhancement layer bitstreams. The base layer bitstream is decoded to produce pixels of a reconstructed base layer image corresponding to the depth image. Each enhancement layer bitstream is decoded in a low to high order to produces a reconstructed residual image. During the decoding of the enhancement layer bitstream, a context model is maintained using an edge map, and each enhancement layer bitstream is entropy decoded using the context model to determine a significance value corresponding to pixels of the reconstructed residual image and a sign bit for each significant pixel, and a pixel value of the reconstructed residual image is reconstructed according to the significance value, sign bit and an uncertainty interval. Then, the reconstructed residual images are added to the reconstructed base layer image to produce the reconstructed depth image.

A method for optic disk detection in retinal images, includes: extracting a vessel tree from a retinal image; locating an optic disk in the retinal image using the vessel tree; enhancing a contrast of the optic disk; removing vessels from the contrast enhanced optic disk; detecting a boundary of the optic disk with vessels removed by generating an edge map; projecting the edge map at a plurality of angles by using a radon transform; and estimating a radius and center of the optic disk using the projections.

The invention relates to a method and a system for extracting field image edges of plant leaves. The method comprises the following steps of: 1, smoothing and denoising a subimage in a partition area adjacent to an area containing a single leaf serving as a target image; 2, performing area-based threshold partition and gradient-based edge detection partition on the image processed by the step 1, and performing initial optimization to acquire an inner template of a target area; 3, judging whether the target area of the image processed by the step 2 is a leaf image or not, if so, executing the step 4, and otherwise, performing grayscale mapping and morphological reconstruction on the image processed by the step 1, and executing the step 2; and 4, performing operation of adhesion disconnection, false edge removal and closure on the image processed by the step 2 to obtain the edge image with a closed edge. The method and the system improve the accuracy and integrality of the extraction ofthe field image edges, which have non-uniform grayscale, of the plant leaves.

The invention belongs to the field of image acquiring and processing, and particularly relates to an image acquiring and processing method based on an FPGA (field programmable gate array) serving as a control core. The method mainly includes building an image acquiring and processing SOPC (system on programmable chip) system comprising a Nios II soft-core processor module, a CMOS (complementary metal-oxide-semiconductor transistor) image sensing controller module, an FIFO (first in first out) controller module, an SDRAM (synchronous dynamic random access memory) controller module, a Canny edge detection algorithm module and a VGA (video graphic array) controller module. Firstly, images are acquired and then cached in the FIFO controller module; secondly, image data in the FIFO controller module are transmitted to an off-chip SDRAM chip by FPGA software and are stored in the off-chip SDRAM chip; thirdly, images in a memory are processed by means of an image processing algorithm, 'ping pong operation' is realized during processing, and the processing images are stored in the SDRAM again; and finally, the acquired edge images are displayed through a VGA interface after being processed.

A method for optic disk detection in retinal images, includes: extracting a vessel tree from a retinal image; locating an optic disk in the retinal image using the vessel tree; enhancing a contrast of the optic disk; removing vessels from the contrast enhanced optic disk; detecting a boundary of the optic disk with vessels removed by generating an edge map; projecting the edge map at a plurality of angles by using a radon transform; and estimating a radius and center of the optic disk using the projections.