Automatic image object identification using threshold gradient magnitude based on terrain type
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06K-009/00
G06K-009/44
출원번호
US-0923502
(2010-09-24)
등록번호
US-8135174
(2012-03-13)
발명자
/ 주소
Wiedemann, Melissa
Phurrough, Larry
Gratsch, Colleen Flynn
Skoblick, Richard
Lee, Harry C.
출원인 / 주소
Lockheed Martin Corporation
대리인 / 주소
Buchanan Ingersoll & Rooney PC
인용정보
피인용 횟수 :
2인용 특허 :
38
초록▼
A method and system for processing image data to identify objects in an image. Terrain types are identified in the image. A second image is generated identifying areas of the image which border regions of different intensities by identifying a gradient magnitude value for each pixel of the image. A
A method and system for processing image data to identify objects in an image. Terrain types are identified in the image. A second image is generated identifying areas of the image which border regions of different intensities by identifying a gradient magnitude value for each pixel of the image. A filtered image is generated from the second image, the filtered image identifying potential objects which have a smaller radius than the size of a filter and a different brightness than background pixels surrounding the potential objects. The second image and the filtered image are compared to identify potential objects as an object. A potential object is identified as an object if the potential object has a gradient magnitude greater than a threshold gradient magnitude, and the threshold gradient magnitude is based on the terrain type identified in the portion of the image where the potential object is located.
대표청구항▼
1. A method for automatically identifying objects in a first image comprising: identifying terrain types in the first image;generating, in a processor of a computer processing device, a second image identifying areas of the first image which border regions of different intensities by identifying a g
1. A method for automatically identifying objects in a first image comprising: identifying terrain types in the first image;generating, in a processor of a computer processing device, a second image identifying areas of the first image which border regions of different intensities by identifying a gradient magnitude value for each pixel of the first image;generating, in the processor, a filtered image from the second image, the filtered image identifying potential objects which have a smaller radius than the size of a filter and a different brightness than background pixels surrounding the potential objects;comparing, in the processor, the second image and the filtered image to identify potential objects as an object, a potential object is identified as an object if the potential object has a gradient magnitude greater than a threshold gradient magnitude, and the threshold gradient magnitude is based on the terrain type identified in the portion of the first image where the potential object is located. 2. A method for automatically identifying objects in a first image comprising: identifying terrain types in the first image;generating, in a processor of a computer processing device, a second image identifying areas of the first image which border regions of different intensities by identifying a gradient magnitude value for each pixel of the first image;generating, in the processor, a filtered image from the second image, the filtered image identifying potential objects which have a smaller radius than the size of a filter and a different brightness than background pixels surrounding the potential objects;comparing, in the processor, the second image and the filtered image to identify potential objects as an object, a potential object is identified as an object if the potential object has a gradient magnitude greater than a threshold gradient magnitude, and the threshold gradient magnitude is based on the terrain type identified in the portion of the first image where the potential object is located; andgenerating a mean local gradient magnitude image using the second image, wherein the threshold gradient magnitude is determined using the mean local gradient magnitude image. 3. The method of claim 2, wherein the threshold comprises a first threshold based on the mean local gradient magnitude, and a second threshold based on a maximum histogram value of a local gradient magnitude. 4. A method for automatically identifying objects in a first image comprising: identifying terrain types in the first image;generating, in a processor of a computer processing device, a second image identifying areas of the first image which border regions of different intensities by identifying a gradient magnitude value for each pixel of the first image;generating, in the processor, a filtered image from the second image, the filtered image identifying potential objects which have a smaller radius than the size of a filter and a different brightness than background pixels surrounding the potential objects;comparing, in the processor, the second image and the filtered image to identify potential objects as an object, a potential object is identified as an object if the potential object has a gradient magnitude greater than a threshold gradient magnitude, and the threshold gradient magnitude is based on the terrain type identified in the portion of the first image where the potential object is located,wherein the step of generating a filtered image comprises: performing a series of dilations and erosions of the second image to produce a spatially filtered image; andsubtracting said spatially filtered image from the first image to produce the filtered image. 5. The method of claim 4, further comprising: performing an intensity histogram of the filtered image to produce a maximum intensity value and a minimum intensity value;generating a minimum brightness value and a maximum brightness value for each terrain type identified in the first image; andremoving portions of the image which are not greater than the maximum intensity value or less than the minimum intensity value, and which are not greater than the minimum brightness value for a particular terrain type for the portion of the first image or less than the maximum brightness value for a particular terrain type for the portion of the first image. 6. A non-transitory computer-readable recording medium having a computer program recorded thereon that causes a computer to identify objects in a first image, the program causing the computer to perform operations comprising: identifying terrain types in the first image;generating a second image identifying areas of the first image which border regions of different intensities by identifying a gradient magnitude value for each pixel of the first image;generating a filtered image from the second image, the filtered image identifying potential objects which have a smaller radius than the size of a filter and a different brightness than background pixels surrounding the potential objects; andcomparing the second image and the filtered image to identify potential objects as an object, a potential object is identified as an object if the potential object has a gradient magnitude greater than a threshold gradient magnitude, and the threshold gradient magnitude is based on the terrain type identified in the portion of the first image where the potential object is located. 7. A non-transitory computer-readable recording medium having a computer program recorded thereon that causes a computer to identify objects in a first image, the program causing the computer to perform operations comprising: identifying terrain types in the first image;generating a second image identifying areas of the first image which border regions of different intensities by identifying a gradient magnitude value for each pixel of the first image;generating a filtered image from the second image, the filtered image identifying potential objects which have a smaller radius than the size of a filter and a different brightness than background pixels surrounding the potential objects;comparing the second image and the filtered image to identify potential objects as an object, a potential object is identified as an object if the potential object has a gradient magnitude greater than a threshold gradient magnitude, and the threshold gradient magnitude is based on the terrain type identified in the portion of the first image where the potential object is located; andgenerating a mean local gradient magnitude image using the second image, wherein the threshold gradient magnitude is determined using the mean local gradient magnitude image. 8. The computer-readable recording medium of claim 7, wherein the threshold comprises a first threshold based on the mean local gradient magnitude, and a second threshold based on a maximum histogram value of a local gradient magnitude. 9. A non-transitory computer-readable recording medium having a computer program recorded thereon that causes a computer to identify objects in a first image, the program causing the computer to perform operations comprising: identifying terrain types in the first image;generating a second image identifying areas of the first image which border regions of different intensities by identifying a gradient magnitude value for each pixel of the first image;generating a filtered image from the second image, the filtered image identifying potential objects which have a smaller radius than the size of a filter and a different brightness than background pixels surrounding the potential objects; andcomparing the second image and the filtered image to identify potential objects as an object, a potential object is identified as an object if the potential object has a gradient magnitude greater than a threshold gradient magnitude, and the threshold gradient magnitude is based on the terrain type identified in the portion of the first image where the potential object is located,wherein the operation of generating a filtered image comprises: performing a series of dilations and erosions of the second image to produce a spatially filtered image; andsubtracting said spatially filtered image from the first image to produce the filtered image. 10. The computer-readable recording medium of claim 9, wherein the operations further comprise: performing an intensity histogram of the filtered image to produce a maximum intensity value and a minimum intensity value;generating a minimum brightness value and a maximum brightness value for each terrain type identified in the first image; andremoving portions of the first image which are not greater than the maximum intensity value or less than the minimum intensity value, and which are not greater than the minimum brightness value for a particular terrain type for the portion of the first image or less than the maximum brightness value for a particular terrain type for the portion of the first image.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (38)
Gallegos Sonia C. (122 Cedarwood Dr. Slidell LA 70461) Holyer Ronald J. (1640 Harbor Dr. ; No. 111 Slidell LA 70458) Peckinpaugh Sarah H. (P.O. Box 1018 Ocean Springs MS 39564) Cheng Chiu-Fu (1110 Hi, Apparatus and method for cloud masking.
Simanovsky Sergey ; Bechwati Ibrahim M. ; Hiraoglu Muzzafer ; Crawford Carl R., Apparatus and method for detecting concealed objects in computed tomography data.
Sato Jun (Cambridge GB2) Saiki Mitsuyoshi (Tsukuba JPX), Apparatus which detects lines approximating an image by repeatedly narrowing an area of the image to be analyzed and inc.
Bell David M. (Palm Bay FL) Bryan Joseph K. (West Melbourne FL) Black Stephen B. (Arlington VA), Mechanism for registering digital images obtained from multiple sensors having diverse image collection geometries.
Haley Paul H. (Monroeville PA) Oravec James J. (Plum Boro PA) Petlevich ; Jr. Walter J. (Uniontown PA) Schweizer Philipp F. (Monroeville PA), Method and apparatus for classifying objects in sonar images.
Kumar, Rakesh; Hsu, Stephen Charles; Hanna, Keith; Samarasekera, Supun; Wildes, Richard Patrick; Hirvonen, David James; Klinedinst, Thomas Edward; Lehman, William Brian; Matei, Bodgan; Zhao, Wenyi; L, Method and apparatus for performing geo-spatial registration of imagery.
Borrel Paul (Peekskill NY) Rossignac Jaroslaw R. (Ossining NY), Multi-resolution graphic representation employing at least one simplified model for interactive visualization applicatio.
Heine John J. ; Clarke Laurence P. ; Deans Stanley R. ; Stauduhar Richard Paul ; Cullers David Kent, Normal and abnormal tissue identification system and method for medical images such as digital mammograms.
Eppler Walter G., System and method for automatically determining the position of landmarks in digitized images derived from a satellite-based imaging system.
Jain Anil K. ; Hong Lin ; Bolle Rudolf Maarten ; Pankanti Sharathchandra Umapathirao, System and method for identifying foreground and background portions of digitized images.
Reis James J. (La Palma CA) Luk Anthony L. (Rancho Palos Verdes CA) Lucero Antonio B. (Anaheim CA) Garber David D. (Cypress CA), Target acquisition and tracking system.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.