Image processing method, electronic device and storage medium

Abstract

The invention provides a picture processing method, comprising: training a sample picture library to obtain a classification function of a sample picture with a marked frame and a sample picture without a marked frame; calculating the classification function of the original picture based on the classification function class value, if the classification value satisfies the first preset condition, it is determined that the original picture includes a marked frame drawn with the preset color; the original picture is converted from RGB space to HSV space to obtain an HSV picture; according to the The value range of the preset color in the HSV space is set, the binarization threshold is set, and each pixel in the HSV picture is binarized to obtain a binarized picture; the target contour is extracted from the binarized picture The position information of the original picture is cropped according to the position information to obtain the partial picture corresponding to the marked frame. The invention also provides an electronic device and a storage medium. The invention can quickly and accurately cut out the partial picture corresponding to the marked frame from the original picture.

Description

technical field [0001] The invention relates to the technical field of computers, in particular to a picture processing method, an electronic device and a storage medium. Background technique [0002] With the development of Internet technology, various network resources are very rich, which greatly facilitates users' acquisition of resources. For example, there are a large number of high-definition landscape pictures on the Internet, so that users may find the pictures they need without taking pictures in person. Among them, in order to mark the local content in the picture that needs attention, some pictures have a marked frame drawn in a certain color. For example, in the photos taken by traffic monitoring, people or vehicles that violate the rules are marked in the photos with red squares, etc. , and these partial content are exactly what users expect to find. Therefore, the user not only hopes to find the desired original picture, but also hopes to obtain the partial ...

AI Technical Summary

Problems solved by technology

However, for such original pictures with marked boxes, since the position information of the marked boxes cannot be directly extracted, it is difficult to precisely crop the original picture to obtain the local pictures corresponding to the marked boxes

Images