50 results about "Seam carving" patented technology

What is disclosed is a method for iterative seam selection in an image resizing system utilizing a seam carving technique. In one embodiment, an importance map is generated for a received source image. Seams are carved through the image from one edge to an opposite edge. An energy is computed for each seam based on pixel importance values. A distance is computed from each seam to a previously selected seam. A weighting for each seam is computed using a defined weighting function and the calculated seam distances. The weighting is applied to the energy of each seam produce a revised energy for each seam. A seam is selected based on the produced revised energy. The image is resized at a location of the selected seam. The process repeats until the image has been resized to a desired target output dimension. In such a manner, unnatural image resizing results are avoided.

A method for resizing an input digital image to form an output digital image with an output aspect ratio, comprising: determining a number of rows or columns that need to be reduced from the input digital image; determining an image energy map for the input digital image; repeatedly determining a seam path responsive to the image energy map and removing pixels along the determined seam path to determine the output digital image, wherein the determined seam path satisfies a constraint that a directional image gradient is less than a gradient threshold for each pixel in the seam path, until either the determined number of rows or columns has been reduced or no valid seam path can be found; and cropping or scaling the output digital image to the output aspect ratio if the determined number of rows or columns was not reduced.

A method for modifying an input digital image having input dimensions defined by a number of input rows and input columns to form an output digital image where the number of rows or columns is reduced by one, comprising an image energy map determined from the input image; determining a seam path responsive to the image energy map; imposing constraints on the seam path; and removing pixels along the seam path to modify the input digital image.

The invention belongs to the field of panoramic visible light image processing, and specifically relates to a panoramic sea-sky line extracting method based on improved Seam Carving. The method comprises the following steps: (1) acquiring the visual salient map of a panoramic sea image; (2) extracting a salient region by use of a dynamic threshold method; (3) constructing an energy function; (4) searching the highest energy line; (5) screening fitting points on the sea-sky line; and (6) fitting a panoramic sea-sky line by use of the least square method. The method is successfully applied to panoramic sea-sky line detection by constructing a new energy function and setting the screening conditions of the points on the sea-sky line, and accurate extraction of a circular sea-sky line in a panoramic visible light image against a complex sea-sky background under different sea conditions and weather conditions is realized.

Systems, methods, and computer-readable storage media for resizing images using seam carving techniques are disclosed. The methods may facilitate efficient re-computation of the energy of an image between seam carving iterations in a resizing operation by re-computing only the energy of pixels and/or seams in the neighborhood of removed and/or replicated seams. The neighborhood and pixels for which energy values are updated may be dependent on the energy function employed by the seam carving techniques. The efficient re-computation of image energy may facilitate the use of seam carving techniques in user-interactive environments. The methods may also be used to pre-compute a retargeting matrix, usable in performing retargeting operations on an input image without re-computing the energy and/or the lowest cost seam(s) of the image between iterations. The methods may be implemented as stand-alone applications or as program instructions implementing components of a graphics application, executable by a CPU and/or GPU.

A system and method for resizing a digitally represented color image are presented. A color image with pixels defined by luminance and at least one chrominance value is received. For each pixel of the color image, a luminance spatial variation and respective chrominance spatial variations in the respective neighborhood of the each pixel are computed. The luminance spatial variation and the respective chrominance spatial variations are combined to produce a respective importance value for each pixel. Selected pixels are identified based upon their respective importance values and are removed by seam carving of the color image. The seam carving identifies seams of pixels based upon the respective importance values of pixels within the seams of pixels to create a resized color image. The resized color image is produced to an image output device.

Systems, methods, and computer-readable storage media for resizing images using seam carving techniques may include generation of a partial solution matrix by at least partially isolating dependencies between sub-problems of a dynamic programming problem corresponding to its solution within different regions of an input image. The number and/or shape of the isolated (or partially isolated) sub-problems may be dependent on the access pattern used by a dynamic programming operation to identify seams in the input image. Multiple sub-problems may be processed independently and in parallel on respective processor core(s) or threads thereof to generate the partial solution matrix. The partial solution matrix may then be processed to identify one or more low-cost seams of the input image. The methods may be implemented as stand-alone applications or as program instructions implementing components of a graphics application, executable by a CPU and/or GPU configured for parallel processing.

The invention relates to a fidelity evaluation method based on the scale adjustment of content relevant images, and pertains to image fidelity evaluation technology. The invention is characterized in that: the evaluation method comprises the steps as follows: A. content relevant scale zooming is carried out to a pre-selected image by an energy function needed to be evaluated and through a seam carving algorithm; B. the corresponding points of the original image are searched in the adjusted image; C. the adjusted image is compared with the original image, and the evaluation result of the fidelity of the energy function is obtained according to the comparative results. The invention also discloses an image fidelity evaluation system for different energy functions by the seam carving algorithm, and the fidelity evaluation system comprises a reference image database, a seam carving algorithm unit and a fidelity evaluation unit. The technical proposal of the invention can evaluate the fidelity of the content relevant scale changed image, thus realizing the fidelity evaluation of the content relevant scale changed image under different energy functions of the image.

The invention discloses a quick seam carving method based on strips and adjacency, and belongs to the field of images. A quick seam carving method is improved and constrained by combining the thought of the adjacency and the strips, so that seams can be uniformly distributed in each area of an image, and a satisfactory result is achieved. The method comprises the following steps of: dividing the image into a plurality of strips at equal intervals, calculating the average importance of the strips by using a significant diagram corresponding to the image, obtaining the target size of each strip according to the importance of the strip and the target size of the whole image by solving an optimization problem with a constraint condition, and acquiring a target size image by adopting the quick seam carving method combining the adjacency in each strip. The calculation speed of the method is further higher than that of other seam carving methods, and the quality of the obtained result image is better than that of the conventional quick seam carving method, so the method has certain application value and significance.

The invention discloses a method for adaptively displaying an image by taking a visual transfer mechanism as a guide, which comprises the following steps of: (1) constructing a visual attention model to obtain a final saliency map; (2) based on the saliency map, sequentially finding points with maximum saliency values from the saliency map according to a competitive mechanism; and (3) determining a final display strategy according to the size of a target saliency screen, a transfer sequence of attention focuses and distances among the attention focuses. By the method for adaptively displaying the image by taking the visual transfer mechanism as the guide and further regulating the accumulated energy of the saliency map after the display strategy is determined, an important reference is provided for the conventional adaptive method for performing cropping and seam carving only according to the accumulated energy, so that a user can fully understand image contents in an adaptive process, and less content loss and better user browsing experiences can be realized for the image with a plurality of salient regions.

As image source attribution techniques have become significantly sophisticated and are now becoming commonplace, there is a growing need for capabilities to anonymize images and videos. Focusing on the photo response non-uniformity noise pattern based sensor fingerprinting technique, this work evaluates the effectiveness of well-established seam carving method to defend against sensor fingerprint matching. We consider ways in which seam-carving based anonymization can be countered and propose enhancements over conventional seam carving method. Our results show that applying geometrical distortion in addition to seam carving will make counter attack very ineffective both in terms of computational complexity and accuracy.

The embodiment of the invention provides a method, a device and a system for image adjustment, which relate to the technical field of multimedia, broaden the zooming range of multimedia resources and can prevent the problem that multimedia resources are distorted or deformed when the zooming range is wide. The embodiment of the invention includes the following steps: the target adjustment value of an image to be displayed is obtained, the seam carving algorithm is utilized to obtain the redundant information, an intermediate image and the like of the image to be displayed, and the shift map algorithm is utilized to zoom out the image to be displayed; in the process of zooming in the image, the intermediate image needs to be zoomed in first after being obtained, and the shift map algorithm is then used for zooming out the zoomed-in image. The invention is mainly applied in the process of zooming multimedia resources.

The invention provides a seam-carving image positioning evidence obtaining method for object deletion and belongs to the field of image blind evidence obtaining in image processing. The seam-carving image positioning evidence obtaining method for object deletion comprises the steps of (1), carrying out block effect grid extraction on a to-be-detected image, thereby obtaining a block effect grid diagram G of the to-be-detected image; (2), carrying out feature extraction on the block effect grid diagram G, thereby obtaining features; (3), classifying the features through utilization of an unsupervised clustering method, thereby obtaining a classification mark result diagram; and (4), carrying out positioning detection the classification mark result diagram. Through application of the method provided by the invention, whether an object in the image is deleted or not can be detected, and the specific position of the deleted object can be positioned.

The present invention discloses a stereoscopic seam carving and scaling method in the vertical direction. The method comprises the steps of A, inputting the left view, the right view and the disparity image of an original stereoscopic image, and setting the resolution of a target image; B, according to an energy function, calculating the energy value of each pixel point in the left view of the stereoscopic image; C, based on the dynamic programming method, finding out a plane seam in the left view of the stereoscopic image in the horizontal direction; D, executing the real projection strategy, and mapping the plane seam in the left view of the stereoscopic image into the right view of the stereoscopic image; E, according to an obtained seam section in the right view of the stereoscopic image through the real projection process, executing a policy based on the combination of the traditional seam carving method and the virtual projection method, spicing up spaced seams to obtain a corresponding seam in the right view of the stereoscopic image in the horizontal direction; F, simultaneously increasing or deleting the seam in the left view of the stereoscopic image and the seam in the right view of the stereoscopic image; G, repeatedly performing the steps C to F till the stereoscopic image is scaled to a specified size, and outputting a final left view and a final right view. According to the technical scheme of the invention, the overall features of the scaled image can be maintained, and important objects are highlighted. Meanwhile, the stereoscopic information of the stereoscopic image can be maintained.