IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0786775
(2004-02-25)
|
등록번호 |
US-7321668
(2008-01-22)
|
우선권정보 |
JP-2003-397999(2003-11-27) |
발명자
/ 주소 |
- Horie,Daisaku
- Nakano,Yuusuke
|
출원인 / 주소 |
- Konica Minolta Holdings, Inc.
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
9 인용 특허 :
7 |
초록
▼
An object detection apparatus is provided that detects a target object correctly without decreasing a process speed so much even if there is a change in a shooting environment. The human body detection apparatus for detecting a human body in an image includes an image input portion for entering an i
An object detection apparatus is provided that detects a target object correctly without decreasing a process speed so much even if there is a change in a shooting environment. The human body detection apparatus for detecting a human body in an image includes an image input portion for entering an image taken by a camera, a plurality of plane generation portions for detecting features of the image by using different methods, a section for obtaining conditions about shooting by the camera, a section for calculating reliability values of features detected by the plane generation portions in the case of the obtained condition, and a detection process portion for detecting a human body in the shot image in accordance with features detected from the shot image detected by one or more of the plane generation portions and reliability values of the features.
대표청구항
▼
What is claimed is: 1. An object detection apparatus for detecting a target object in an image, comprising: an image input portion for entering a shot image that is taken by a camera; a plurality of feature detection portions for detecting features of the shot image by using different methods; a sh
What is claimed is: 1. An object detection apparatus for detecting a target object in an image, comprising: an image input portion for entering a shot image that is taken by a camera; a plurality of feature detection portions for detecting features of the shot image by using different methods; a shooting condition obtaining portion for obtaining information indicating conditions for shooting by the camera; a reliability calculation portion for calculating reliability of the feature that is detected by each of the feature detection portions in the conditions indicated by the information obtained by the shooting condition obtaining portion; and an object detection portion for detecting the object in the shot image in accordance with the features detected respectively by one or more of the plural feature detection portions from the shot image and the reliability of the features calculated by the reliability calculation portion. 2. The object detection apparatus according to claim 1, wherein the feature detection portion detects the feature as a feature quantity, and the object detection portion detects the object in accordance with the feature quantity of each feature that has a weight corresponding to the reliability of each feature. 3. The object detection apparatus according to claim 2, wherein the feature detection portion delivers a feature image that indicates a feature degree indicating the degree of the feature in each pixel of the shot image as a result of detecting the feature quantity of the shot image, and the object detection portion detects the object in accordance with the feature image. 4. The object detection apparatus according to claim 3, further comprising a composite image generation portion for generating a composite image by adding values of corresponding pixels of plural feature images, wherein the feature detection portion delivers the feature image for each of the plural sorts of features, the composite image generation portion generates the composite image in accordance with each of the plural sorts of feature images, and the object detection portion detects the object in accordance with the composite image generated by the composite image generation portion. 5. The object detection apparatus according to claim 4, wherein the object detection portion detects a position of the object in accordance with a pixel having a pixel value larger than a predetermined value and pixel values of pixels surrounding said pixel among the pixels of the composite image. 6. The object detection apparatus according to claim 1, further comprising a reliability memory portion for memorizing the reliability of the feature calculated by the reliability calculation portion, wherein the reliability calculation portion calculates the reliability of the feature at a predetermined timing, and the object detection portion detects the object in the shot image in accordance with the latest feature memorized in the reliability memory portion. 7. The object detection apparatus according to claim 1, further comprising a shooting condition memory portion for memorizing information that indicates the conditions obtained by the shooting condition obtaining portion, wherein the reliability calculation portion performs a process for calculating the reliability if a difference between the information indicating the conditions obtained by the shooting condition obtaining portion and the information that is memorized in the shooting condition memory portion and indicates the conditions in the past is larger than a predetermined quantity or a predetermined ratio. 8. The object detection apparatus according to claim 1, wherein the object is a human body, and the feature detection portion is a section for calculating a matching degree between the shot image and a template having a semiellipse shape for obtaining the feature, a section for detecting a likelihood of a flesh color in each of sectioned areas of a pixel plane of the shot image for obtaining the feature, a section for detecting a likelihood of a hair color in the area of the shot image for obtaining the feature, or a section for calculating a matching degree between the shot image and a template having shapes of shoulders for obtaining the feature. 9. The object detection apparatus according to claim 1, wherein the shooting condition obtaining portion obtains information about setting of the camera, information about a state of a shooting area of the camera, or information about an object of which the camera takes an image as the information that indicates the conditions. 10. An object detection apparatus for detecting a target object in an image, comprising: an image input portion for entering a shot image that is taken by a camera; a plurality of feature detection portions for detecting features of the shot image by using different methods; a shooting condition obtaining portion for obtaining information indicating conditions for shooting by the camera; a reliability calculation portion for calculating reliability of the feature that is detected by each of the feature detection portions in the conditions; an operation method decision portion for deciding an operation method for detecting the object in accordance with the reliability calculated by the reliability calculation portion for each feature detected by each of the feature detection portions; and an object detection portion for detecting the object in the shot image in accordance with the features detected respectively by one or more of the plural feature detection portions and the operation method decided by the operation method decision portion. 11. The object detection apparatus according to claim 10, wherein the feature detection portion delivers a feature image that indicates a feature degree indicating the degree of the feature in each pixel of the shot image as a result of detecting the feature quantity of the shot image, and the object detection portion detects the object in accordance with the feature image. 12. The object detection apparatus according to claim 11, further comprising a composite image generation portion for generating a composite image of plural feature images by performing an operation in accordance with the operation method decided by the operation method decision portion, wherein the feature detection portion delivers the feature image for each of the plural sorts of features, the composite image generation portion generates the composite image in accordance with each of the plural sorts of feature images, and the object detection portion detects the object in accordance with the composite image generated by the composite image generation portion. 13. The object detection apparatus according to claim 12, wherein the object detection portion detects a position of the object in accordance with a pixel having a pixel value larger than a predetermined value and pixel values of pixels surrounding said pixel among the pixels of the composite image. 14. The object detection apparatus according to claim 10, further comprising a shooting condition memory portion for memorizing information that indicates the conditions obtained by the shooting condition obtaining portion, wherein the reliability calculation portion performs a process for calculating the reliability if a difference between the information indicating the conditions obtained by the shooting condition obtaining portion and the information that is memorized in the shooting condition memory portion and indicates the conditions in the past is larger than a predetermined quantity or a predetermined ratio. 15. The object detection apparatus according to claim 10, wherein the shooting condition obtaining portion obtains information about setting of the camera, information about a state of a shooting area of the camera, or information about an object of which the camera takes an image as the information that indicates the conditions. 16. An object detection method for detecting a target object in an image, comprising: a step of entering a shot image that is taken by a camera; a step of detecting features of the shot image by using different feature detection methods; a step of obtaining information indicating conditions for shooting by the camera; a step of calculating reliability of the feature that is detected by each of the feature detection methods in the conditions indicated by the obtained information; and a step of detecting the object in the shot image in accordance with the features detected respectively by one or more of the plural feature detection methods from the shot image and reliability of the features. 17. The object detection method according to claim 16, wherein the step of detecting the feature includes detecting the feature as a feature quantity, and the step of detecting the object includes detecting the object in accordance with the feature quantity of each feature that has a weight corresponding to the reliability of each feature. 18. The object detection method according to claim 16, further comprising a step of deciding an operation method for detecting the object in accordance with the reliability of each of the features, wherein the step of detecting the object includes detecting the object in the shot image in accordance with the operation method decided on the basis of the reliability of the feature. 19. A computer-readable medium having stored a computer program product for performing the steps of: entering a shot image that is taken by a camera; detecting features of the shot image by using different feature detection methods; obtaining information indicating conditions for shooting by the camera; calculating reliability of the feature that is detected by each of the feature detection methods in the conditions indicated by the obtained information; and detecting the object in the shot image in accordance with the features detected respectively by one or more of the plural feature detection methods from the shot image and the reliability of the features. 20. A monitoring system comprising: a video camera for taking an image; and an object detection apparatus for detecting a target object in the image taken by the video camera, including an image input portion for entering a shot image that is taken by the video camera, a plurality of feature detection portions for detecting features of the shot image by using different methods, a shooting condition obtaining portion for obtaining information indicating conditions for shooting by the video camera, a reliability calculation portion for calculating reliability of the feature that is detected by each of the feature detection portions in the conditions indicated by the information obtained by the shooting condition obtaining portion, and an object detection portion for detecting the object in the shot image in accordance with the features detected respectively by one or more of the plural feature detection portions from the shot image and the reliability of the features calculated by the reliability calculation portion. 21. The monitoring system according to claim 20, further comprising an operation method decision portion for deciding an operation method for detecting the object in accordance with the reliability calculated by the reliability calculation portion, of each feature detected by each of the feature detection portions, wherein the object detection portion detects the object in the shot image in accordance the operation method determined on the basis of the reliability of the feature. 22. The monitoring system according to claim 20, further comprising an image display device for display an image area of the object detected by the object detection apparatus after enlarging the image area within the image taken by the video camera. 23. The monitoring system according to claim 20, further comprising a recording device for recording the image if the object is detected in the image taken by the video camera.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.