Method for processing an image
An image and key point technology, applied in the field of computer vision, can solve the problem of building a tree consuming too many bits, performance degradation, etc., to achieve the effect of efficient analysis and representation, improved compression rate, and limited complexity
Patent Information
- Authority / Receiving Office
- CN · China
- Current Assignee / Owner
- Publication Date
- 2014-08-27
Smart Images
Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] The present invention relates to image processing techniques in the field of computer vision, and in particular to the subject commonly known as visual search or augmented reality. In visual search and augmented reality applications, information extracted from an image or image sequence is sent to a server where it is compared with information extracted from a database of reference images or image sequences. A reference image or sequence of images represents a model of the object to be recognized. In this context, the invention relates to the compression of information extracted from an image or sequence of images sent to a server, in particular to the compression of the positions of points of interest extracted from an image or sequence of images. Background technique
[0002] Visual Search (VS) is an automated system that identifies an object or objects depicted in an image or sequence of images by analyzing only the visible portion of the image or ...
Examples
Embodiment Construction
[0101] figure 1 Shown is a schematic diagram of an image processing method 100 according to an embodiment. The image processing method 100 includes: providing (101) a keypoint set from an image; describing (103) position information of the keypoint set in the form of a binary matrix; and scanning (105) the binary matrix according to a predetermined order, thereby creating a position of the keypoint set A new representation of information. In an implementation form, the form of the new representation of the position information of the keypoint set is another binary matrix.
[0102] figure 2 Shown is a schematic diagram of a location information compression method 201 according to an implementation form. The image compression method 201 includes generating (200) a histogram and a histogram count, compressing (210) the histogram, compressing (220) the histogram count, and creating (240) an encoded bitstream from the compressed descriptor 230. Histogram based on Figure 16 ...