Blocking dimension based image library establishing method

Abstract

The invention discloses a blocking dimension based image library establishing method and belongs to the technical field of image processing. The blocking dimension based image library establishing method comprises the steps that images with different blocking targets are acquired, and the acquired images are classified according to the blocking targets to form a tree structure; each image is marked according to blocking dimensions; the marked images are added to corresponding tree-like classification structures so as to form an image library; follow-up acquired blocking images are successively added to the image library by adopting the same processing method to further update and perfect the image library. The blocking dimension based image library establishing method has the advantages of being strong in pertinence, good in fundamentality and wide in application prospect.

Description

technical field

[0001] The invention belongs to the technical field of image processing, and in particular relates to an image library construction method based on occlusion dimensions. Background technique

[0002] As a medium of information dissemination, images are widely used because they contain intuitive and rich information. In order to promote the development of computer vision, especially the research on image segmentation, target detection, and recognition methods, some international evaluation platforms have emerged to compare and detect the pros and cons of each algorithm: In 2005, the EU established PASCAL (Pattern Analysis, Statistical Modeling and Computational Learning) data set, opened the VOC (Visual Object Classes) challenge; the VOC image library (Visual Object Classes) image set is divided into 4 categories, namely transportation, indoor objects, animals, and others; a total of 20 categories are included The catalog contains a total of 11,530 pictures, ...

Examples

Embodiment Construction

[0015] The present invention will be further described below in conjunction with accompanying drawing and embodiment:

[0016] A method for constructing an image library based on the occlusion dimension proposed by the present invention, such as figure 1 As shown, it specifically includes the following steps:

[0017] 1) Collect images with different occlusion targets, and classify the collected images according to the occlusion targets to form a tree structure;

[0018] 2) Mark each image according to the occlusion dimension;

[0019] 3) Add the marked images to the corresponding tree classification structure to form an image library; add the subsequently collected occlusion graphics to the image library successively according to the processing methods of steps 1) and 2), so that the image library can be further updated and perfect.

[0020] The above steps 1) collect images with different occlusion targets, and classify the collected images according to the occlusion targ...

technical field

[0001] The invention belongs to the technical field of image processing, and in particular relates to an image library construction method based on occlusion dimensions. Background technique

[0002] As a medium of information dissemination, images are widely used because they contain intuitive and rich information. In order to promote the development of computer vision, especially the research on image segmentation, target detection, and recognition methods, some international evaluation platforms have emerged to compare and detect the pros and cons of each algorithm: In 2005, the EU established PASCAL (Pattern Analysis, Statistical Modeling and Computational Learning) data set, opened the VOC (Visual Object Classes) challenge; the VOC image library (Visual Object Classes) image set is divided into 4 categories, namely transportation, indoor objects, animals, and others; a total of 20 categories are included The catalog contains a total of 11,530 pictures, ...

Examples

Embodiment Construction

[0015] The present invention will be further described below in conjunction with accompanying drawing and embodiment:

[0016] A method for constructing an image library based on the occlusion dimension proposed by the present invention, such as figure 1 As shown, it specifically includes the following steps:

[0017] 1) Collect images with different occlusion targets, and classify the collected images according to the occlusion targets to form a tree structure;

[0018] 2) Mark each image according to the occlusion dimension;

[0019] 3) Add the marked images to the corresponding tree classification structure to form an image library; add the subsequently collected occlusion graphics to the image library successively according to the processing methods of steps 1) and 2), so that the image library can be further updated and perfect.

[0020] The above steps 1) collect images with different occlusion targets, and classify the collected images according to the occlusion targ...

Images