Extraction method for region of interest

Abstract

The invention discloses an extraction method for a region of interest. The extraction method specifically comprises the following steps of: 1, preprocessing an original image; 2, calculating a salient map for the original image; 3, cutting the original image by using a watershed segmentation algorithm; 4, calculating interest of each region according to the salient map and a segmentation result; and 5, detecting the region of interest. According to the extraction method, the region of interest in the image can be automatically searched and found; an Itti model for simulating a human attention mechanism is adopted; and the found region of interest is basically matched with the subjective feeling of a person. Compared with manual marking of the region of interest, the extraction method is faster and more accurate.

Description

technical field [0001] The invention belongs to the field of computer image processing, and proposes a method for extracting a region of interest. Background technique [0002] The rapid development of digital images and the Internet has brought about an explosive growth of digital image resources, because images are more expressive and more intuitive than words. However, compared with the expression form of text, images generally occupy a large amount of storage space. How to express the content of images with less storage space has become a crucial issue. Considering that the content of the image is rich and colorful, and users often only pay attention to a certain key area in the picture, the JPEG2000 standard proposes a region-of-interest coding mechanism, that is, the picture can be divided into multiple areas, and different compression rates are used for different areas. to encode. This mechanism can solve the above problems to a certain extent, but there is still an...

AI Technical Summary

Problems solved by technology

At present, the commonly used methods for image segmentation include similarity measurement algorithm, interest measurement detection technology, and algorithm for automatically extracting regions of interest from shallow depth-of-field images. Similarity measurement algorithms are very difficult to extract new features that reflect high-level semantics in images. , the key is that there is a huge gap between the low-level features of the image and the high-level semantics, and the use of inflection points to extract the region of interest of the image has great limitations.

The disadvantage of the inflection point detector is that since the inflection points are concentrated in the texture area, the points of interest extracted by the point of view detector are concentrated in the area with more texture, and the distribution in the area with less texture is very coefficient

This uneven distribution of texture is not conducive to a complete description of the content of each part of the image

Images