Topic
Image conversion
About: Image conversion is a research topic. Over the lifetime, 2490 publications have been published within this topic receiving 19077 citations.
Papers published on a yearly basis
Papers
More filters
Patent•
22 Mar 2017
TL;DR: In this paper, a multidimensional characteristic extraction method and device based on connected domain analysis was proposed, in which an obtained gray scale image was preprocessed to obtain a region of interest, the region was then subjected to first noise reduction to remove noise points, then the region after the first noise reductions was subjected to boundary tracking and marking to obtain the target region, and finally characteristic extraction was carried out.
Abstract: The invention discloses a multidimensional characteristic extraction method and device based on connected domain analysis. An obtained gray scale image is preprocessed to obtain a region of interest, the region of interest is subjected to first noise reduction to remove noise points, then the region of interest after the first noise reduction is subjected to boundary tracking and marking to obtain a target region, the target region is subjected to second noise reduction to remove noise blocks in the target region, and finally characteristic extraction is carried out. The accuracy of the image is improved by two times of noise reduction processing before extracting the characteristics, and the extracted characteristics are enabled to be more accurate; the gray scale image is compressed by run-length coding, and the run-length coding of the gray scale image conversion is operated in the preprocessing, two times of noise reduction and boundary tracking and connected characteristic marking of the gray scale image, so that the operation speed is improved, and processing demands of real-time images are met; and 20 characteristics can be extracted, and industrial field detection demands can be basically met.
1 citations
Patent•
30 Sep 2004
TL;DR: In this article, the area of each component constituting the filter after rotation was divided and included in the area set for each component constituted by a new filter, and then the numerical value of the components was divided on the basis of the ratio, and added a value which corresponds to the area and is obtained by the division.
Abstract: PROBLEM TO BE SOLVED: To more accurately detect components including even an angular component. SOLUTION: A filter generation processing means 1A sets an area for each component constituting a reference filter used for image processing. When a filter which will be a standard is rotated to an angle with a component which will be the center of the filter which will be the standard at the center, computes how will the area of each component constituting the filter after rotation be divided and included in the area set for each component constituting a new filter, for each component constituting the filter after the rotation by a ratio of the area, divides the numerical value of each component on the basis of the ratio, and adds a value which corresponds to the area and is obtained by the division, for each area of each component constituting the new filter to obtain the value of each component constituting the new filter. COPYRIGHT: (C)2004,JPO&NCIPI
1 citations
Patent•
25 Jan 1994
TL;DR: In this paper, the exclusive logical OR operation on an image subjected to morphology processing and an image after labeling is carried out to detect burr automatically in a highly accurate manner without entailing any registration of a nondefective image or precision positioning.
Abstract: PURPOSE:To detect burr automatically in a highly accurate manner without entailing any registration of a nondefective image or precision positioning by carrying out the exclusive OR operation on an image subjected to morphology processing and an image after labeling. CONSTITUTION:An inspected part 2 (O-ring packing) on a conveyor belt 1 is irradiated by means of reflected illumination, photographed with a camera unit 4, and it is taken into an image processor 5 which perform binary coding, on this taken-in image into and this coded image is subjected to labeling for image conversion. Next, the secured image is made into Morphology processing, and it is subjected to logical calculus between images (exclusive logical OR) together with the image after the labeling, thereby detecting any burr in the inspected part 2. Detected data are inputted into a controller 6, and in the case where this burr is so judged that it is to be eliminated, a discharger 7 and a defective recover 8 are operated, thus all defectives are eliminated.
1 citations
Patent•
12 Jan 2012
TL;DR: In this article, the authors propose an imaging apparatus or the like which is capable of capturing a desired subject from the beginning of motion picture capturing when a still picture capturing mode is switched to a motion-picture capturing mode.
Abstract: PROBLEM TO BE SOLVED: To provide an imaging apparatus or the like which is capable of capturing a desired subject from the beginning of motion picture capturing when a still picture capturing mode is switched to a motion picture capturing mode.SOLUTION: An imaging apparatus includes: a subject-of-interest detection section for detecting a subject from an imaging region image that is an image corresponding to image data produced by an imaging section when starting motion picture capturing; a region setting section for initially setting a recording region segmented for recording from the imaging region image in a second aspect ratio based on a position of the subject detected by the subject-of-interest detection section and setting moving a center position of the initially set recording region toward a center position of the imaging region image just by a predetermined amount; and an image segmenting section for producing an image segmented from the imaging region image based on the recording region set by the region setting section.
1 citations
Patent•
05 Jan 2012
TL;DR: In this paper, a video processing method is proposed to enable stereoscopic appearance of an in-plane part of a target object to be naturally enhanced, which is called video enhancement.
Abstract: PROBLEM TO BE SOLVED: To enable stereoscopic appearance of an in-plane part of a target object to be naturally enhanced.SOLUTION: A video processing method comprises at least: frequency decomposition means for performing spatial frequency decomposition to the spatial frequency component of a two-dimensional image to calculate frequency components; counting means for counting each of the frequency components obtained via the frequency decomposition means; determination means for determining a dynamic property of the scene from the result obtained via the counting means; parallax amount designation means for designating a parallax amount to be given to an image to be three-dimensionalized based on the result obtained via the determination means; parallax amount allocation means for allocating the parallax amount to the frequency components obtained by the decomposition; frequency composition means for combining each frequency component obtained via the parallax amount allocation means; image conversion means for outputting a left eye image and a right eye image which enable a binocular stereoscopic vision from the result obtained by the frequency composition means; and the video output means for outputting a video from plural images.
1 citations