专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

JP2000242795A DEVICE AND METHOD FOR DETECTING PICTURE DIFFERENCE AND RECORDING MEDIUM WHERE SAME METHOD IS RECORDED 失效
公开(公告)号：JP2000242795A
公开(公告)日：2000-09-08
申请号：JP4766099
申请日：1999-02-25
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： ARAI HIROYUKI , KUWANO HIDEKATSU , KURAKAKE MASAHARU , SUGIMURA TOSHIAKI
IPC分类号： H04N5/278 , G06T7/20
摘要： PROBLEM TO BE SOLVED: To obtain a method and a device for detecting video difference, which can detect differences between a master video and a derivative video which is generated by partially editing it, and also can detect and classify differences between the different derivative video derived from the same master video. SOLUTION: A standard picture and an objective picture are expanded into series of frame images (S1), the both pictures are made to correspond temporally to each other by detecting and adjusting the time deviation between the frame system of the both picture (S2), and the both pictures are made to correspond spatially to each other by detecting and adjusting the spatial deviation about the frame picture of the both pictures (S3). Whether there is a difference between the frame pictures or not is detected by performing a local picture matching to all the frame pictures of the both pictures (S4) and the time section where frame images whose differences are detected succeed is detected as a differential frame section of the both picture (S5).

2. 发明专利

JP2000194851A CHARACTER AREA EXTRACTION METHOD AND DEVICE AND STORAGE MEDIUM RECORDING THE METHOD 失效
公开(公告)号：JP2000194851A
公开(公告)日：2000-07-14
申请号：JP36654298
申请日：1998-12-24
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： KUWANO HIDEKATSU , ARAI HIROYUKI , KURAKAKE MASAHARU , SUGIMURA TOSHIAKI
IPC分类号： H04N5/278 , G06T7/00 , G06T7/20 , G06T7/40 , H04N9/74
摘要： PROBLEM TO BE SOLVED: To provide a character area extraction method and device for smoothing a contour shape and suppressing extraction failures when a character area is extracted from images including superimposed characters or the like. SOLUTION: Adjacent pixels whose luminance values are similar in color images are turned to a connected area in a luminance similar area formation part 2. In this case, after performing binarization by the line unit of respective horizontal and vertical directions, edge pixels are removed, a character part is extracted separately from a background, binarized results for the respective directions are superimposed, compared and integrated and the connected area is formed. For the character part, since the binarized results for the respective horizontal and vertical directions become common position and shape, the extraction failure of the character area is suppressed. The adjacent pixels whose saturation values are similar in the connected area pixels are formed as the connected area in a saturation value similar area formation part 3. A plurality of pixels adjacent within an image space whose hue values are similar in the connected area pixels whose luminance values and saturation values are both similar are formed as the connected area in a hue value similar area formation part 4. The connected area is defined as a character area image.

3. 发明专利

JP2000187732A METHOD AND DEVICE FOR DECIDING CHARACTER AREA AND RECORDING MEDIUM STORING THE METHOD 失效
公开(公告)号：JP2000187732A
公开(公告)日：2000-07-04
申请号：JP36365198
申请日：1998-12-22
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： KUWANO HIDEKATSU , ARAI HIROYUKI , KURAKAKE MASAHARU , SUGIMURA TOSHIAKI
IPC分类号： G06K9/36 , G06T7/00 , G06T7/40 , H04N5/278 , H04N9/74
摘要： PROBLEM TO BE SOLVED: To improve the discrimination accuracy of each character area in an image including characters. SOLUTION: A color image input storage part 1 inputs color image data including characters and stores the inputted data in a memory. An image area division part 2 divides the inputted and stored color image into connected pixel areas by using a previously determined method. A high contrast area deciding part 3 discriminates a character area highly accurately by executing processing for detecting an edge from an area boundary part between plural divided areas, leaving an area in which the ratio of peripheral length to edge pixels is more than a threshold as a high contrast area having high possibility of including a character and disusing other areas as low contrast areas and then stores the obtained character area image in an character area image storing part 4.

4. 发明专利

JP2000181994A CHARACTER RECOGNITION PROCESSING METHOD, DEVICE THEREFOR AND RECORDING MEDIUM RECORDING THE METHOD 失效
公开(公告)号：JP2000181994A
公开(公告)日：2000-06-30
申请号：JP35707298
申请日：1998-12-16
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MORI MINORU , KURAKAKE MASAHARU , SUGIMURA TOSHIAKI
IPC分类号： G06K9/62 , G06K9/80
摘要： PROBLEM TO BE SOLVED: To provide a processing method and the device for correctly recognizing a character even in a condition where background noise is present. SOLUTION: A character pattern is divided into local areas in a preprocessing part 102 and the number of black pixels present inside the local area is measured for each local area in a black pixel number extraction part 1-4. In an identification part 1-5, a difference value from the standard black pixel number of the respective categories of a prepared black pixel number standard dictionary 1-7 is obtained, and at the time of calculating a distance value or similarity between a feature vector obtained from the local area of an input pattern in a feature extraction part 1-3 and the standard feature vector of the corresponding category of the feature standard dictionary 1-7, adjustment is performed so as to reduce the distance value or the value of the similarity among the respective local areas by the value proportional to the black pixel number difference value. Thus, the fluctuation of the distance value or the similarity in the local area including a lot of noise is reduced, the fluctuation of the distance value or the similarity in the entire character pattern is reduced and correct character recognition is made possible.

5. 发明专利

JPH11239363A DEVICE FOR EXTRACTING CHARACTER AREA IN VIDEO IMAGE, ITS METHOD AND MEDIUM FOR RECORDING ITS METHOD 失效
公开(公告)号：JPH11239363A
公开(公告)日：1999-08-31
申请号：JP3947198
申请日：1998-02-23
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： KUWANO HIDEKATSU , ARAI HIROYUKI , KURAKAKE MASAHARU , OGURA KENJI
IPC分类号： H04N5/91 , G06T7/40 , H04N9/00 , H04N9/74 , H04N9/79
摘要： PROBLEM TO BE SOLVED: To extract a deteriorated color part in characters with high luminance and to suppress extraction of an area with a low contrast with respect to its surrounding areas in the case of extracting a character area from an image. SOLUTION: A plural frames averaging section 2 receives a plurality of color images and obtains a mean value of each of RGB components of pixels corresponding to each other between images to generate a mean color image. A similar luminance value area generating section 3 sets a plurality of adjacent pixels as a connection area whose luminance values are similar in the means value color image. A similar saturation value area generating section 4 sets a plurality of adjacent pixels as a connection area whose saturation values are similar in the connection area in which the luminance is also similar. A similar hue value area generating section 5 sets a plurality of adjacent pixels as a connection area whose hue values are similar in the connection area in which the luminance and the saturation area similar. A background area elimination section 6 eliminates a connection area not satisfying a character area characteristic with respect to the connection area where the luminance value, the saturation value and the hue value are similar to obtain a character area.

6. 发明专利

JPH11175536A VIDEO INFORMATION STRUCTURING PROCESSING METHOD AND DEVICE THEREFOR 失效
公开(公告)号：JPH11175536A
公开(公告)日：1999-07-02
申请号：JP33676697
申请日：1997-12-08
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MATSUMURA SUEKI , ABE SHINJI , KURAKAKE MASAHARU , OGURA KENJI
IPC分类号： G06F17/30 , G06T7/20
摘要： PROBLEM TO BE SOLVED: To reduce the processing load of the construction of structured knowledge information, to improve the readability of the structured knowledge information using videos in various fields as objects, and to attain sharing and re-use, in a structuring processing for giving meaning for video information being the time sequence of an image. SOLUTION: A cut change point is detected from a frame picture sequence in an objective video (12), and various pattern features are extracted from those cut change point pictures (13). Then, the content of the structuring knowledge is analyzed based on the extracted pattern features by using a structured knowledge description script (14) for each video field which can be replaced according to the change of a video field being the object of structuring by describing it according to a prescribed grammatical system, and a correspondence relation between the cut change point image and a time section is generated based on the analyzed result (16).

7. 发明专利

JPH11102439A METHOD AND DEVICE FOR RETRIEVING OBJECT IN IMAGE 失效
公开(公告)号：JPH11102439A
公开(公告)日：1999-04-13
申请号：JP26297997
申请日：1997-09-29
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： ARAI HIROYUKI , KURAKAKE MASAHARU , OGURA KENJI
IPC分类号： G06T7/00 , G06T1/00
摘要： PROBLEM TO BE SOLVED: To provide a method and a device which can retrieve an object in an image with high accuracy. SOLUTION: First, a timewise distribution of colors in a retrieval key image of an object is examined and it is decomposed into plural parts (1). Next, the retrieval key image and a retrieval object image are divided into blocks, and a color histogram for each block is calculated (2). Then, a part image that shows the certainty in which each point in the retrieval key image and the retrieval object image belonging to each of the parts is produced by color histogram matching that reflects color similarity (3). An adjacent histogram that represents arrangement relation of the parts is calculated by using the part image (4). Next, the certainty value of the part image which is created from the retrieval object image is updated by using an adjacent histogram, thus the effects of background is reduced (5). Finally, the existence of the object in the retrieval object image is discriminated (7) by collating the adjacent histogram (6) that is calculated from the retrieval key image and a part image of the updated retrieval object image.

8. 发明专利

JPH10301948A METHOD FOR RETRIEVING IMAGE AND VIDEO 失效
公开(公告)号：JPH10301948A
公开(公告)日：1998-11-13
申请号：JP10870197
申请日：1997-04-25
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： ARAI HIROYUKI , KURAKAKE MASAHARU , ODAKA KAZUMI
IPC分类号： G06F17/30 , G06T1/00
摘要： PROBLEM TO BE SOLVED: To attain more accurate subject retrieval and stable subject retrieval independent of the size of each subject. SOLUTION: An image example presented as a retrieving key is decomposed to plural components by checking the spatial distribution of colors in the image example (1). Then the image example and an image to be retrieved are respectively divided into plural rough blocks and color histograms in respective blocks are calculated (2). As to the image example and the image to be retrieved, color histograms in respective blocks are compared with respective components, blocks constituting a certain component are extracted and hierarchically superposed to generate a component image indicating an area constituting the component (3). Then adjacent histograms expressing the arrangement relation of components as histograms are calculated by using these component images (4). Finally the adjacent histograms obtained from the image example are collated with the adjacent histograms obtained from the image to be retrieved to judge the existence of a subject in the image to be retrieved (5).

9. 发明专利

JPH02236694A HANDWRITTEN CHARACTER RECOGNIZING METHOD 失效
公开(公告)号：JPH02236694A
公开(公告)日：1990-09-19
申请号：JP5618189
申请日：1989-03-10
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： KURAKAKE MASAHARU
IPC分类号： G06K9/62
摘要： PURPOSE:To reflect the relation among characteristics as many as a statistically sufficient number on a recognizing processing and recognizing the relation by using a generation model. CONSTITUTION:A distance calculating part 4, which compares a recognition objective character pattern with a non-linear generated model at every character category of a memory 3, of a generated model structure and a parameter, that stores the non-linear generated model structure from plural samples at every character category by obtaining the parameter beforehand, and calculates the distance, is provided. Further the form of the generated model is obtained by combining a non-linear mean pattern reference type and a non-linear self- regression type, and based on the data obtained at every character category beforehand, an optimal structure/parameter value is determined by a statistic evaluation reference. Further the distance of the recognition objective character pattern up to each character category is defined as the deviation of each character category from the non-linear generated model. Thus the character can be recognized in the correct character category.

10. 发明专利

JP2000067162A METHOD AND DEVICE FOR RECOGNIZING CHARACTER PATTERN AND RECORD MEDIUM WHERE SAME METHOD IS RECORDED 失效
公开(公告)号：JP2000067162A
公开(公告)日：2000-03-03
申请号：JP23252298
申请日：1998-08-19
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MORI MINORU , KURAKAKE MASAHARU , SUGIMURA TOSHIAKI
IPC分类号： G06K9/46 , G06K9/62
摘要： PROBLEM TO BE SOLVED: To provide the method and device which recognize a binarized character pattern well even in the case of a position shift of a character line and deterioration in the shape of an outline part. SOLUTION: For a character pattern which is binarized into black and white, a preprocess part 1-2 performs a process for dividing the pattern into areas by characters and a normalizing process for the positions and sizes of the characters. A feature extraction part 1-3 divides the character pattern of one divided and normalized character into mesh areas, counts the connection numbers of white pixels present in each mesh area in plural directions from the white pixels, accumulates by the mesh areas the values of the degree of direction contribution obtained by averaging the white pixel connection numbers by the directions with the values obtained by accumulating the count results, and averages the accumulated values by the white pixel numbers of the respective mesh areas to obtain feature values regarding the two-dimensional structure of the characters. The feature values represents the mutual arrangement relation of character lines etc., so they are used by a discrimination part 1-4 to performs character recognition for the character pattern.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式