Granted Patent B1

US 12,610,136 · App. 19/233,121 · Granted Apr 21, 2026

Inventors: Qinzhang Wu (Chengdu, CN); Xiaoyin Hu (Chengdu, CN); Xiuwen Wu (Chengdu, CN)

Assignee: SICHUAN KANGJISHENG TECHNOLOGY CO., LTD

H04N23/651G06V10/40

Abstract

Claims (24)

1 . A visual image processing method based on a strain mechanism, comprising:

acquiring a plurality of pieces of photon energy information based on a CCD camera or an infrared camera, and extracting variable elements between photons based on the photon energy information, wherein the variable elements comprise a fluctuation difference variation, a dimensionality variation and a frequency variation; wherein, acquiring a plurality of pieces of photon energy information based on a CCD camera or an infrared camera comprises: acquiring a plurality of pieces of photon energy information by the CCD camera or the infrared camera, and converting the plurality of pieces of photon energy information into an electronic signal; and generating, by the CCD camera or the infrared camera, a target image corresponding to the plurality of pieces of photon energy information; extracting variable elements between photons based on the photon energy information comprises: extracting fluctuation differences between photon energy of pixels in each row of pixels in the target image based on a preset detection algorithm; performing division according to a plurality of fluctuation differences in each row to obtain a plurality of energy segments; and acquiring a frequency of each energy segment, and determining an initial spatial position, an end spatial position and a length corresponding to the energy segment;

generating a plurality of energy fields according to the variable elements between the photons, wherein each of the energy fields comprises a plurality of pieces of feature information, and types of the feature information comprise boundary features, form features and scale features; wherein, generating a plurality of energy fields according to the variable elements between the photons comprises: combining energy segments that have the same photon energy and are adjacent to each other according to the initial spatial positions, the end spatial positions and the lengths corresponding to the plurality of energy segments, to obtain a plurality of energy fields; and performing boundary information extraction on the energy fields to obtain feature information corresponding to the energy fields; and

performing spatial reasoning according to the feature information of the energy fields, and splicing the energy fields to obtain a geometric shape of an object, wherein performing spatial reasoning according to the feature information of the energy fields and splicing the energy fields to obtain a geometric shape of an object comprise: performing spatial adjacent splicing according to form features of the energy fields to obtain a body form; performing geometric feature analysis on the body form to obtain a geometric center; constructing geometric framework segments based on the geometric center; and

obtaining the geometric shape of the object in the target image according to the geometric framework segments.

2 . The visual image processing method based on a strain mechanism according to claim 1 , wherein performing division according to a plurality of fluctuation differences in each row to obtain a plurality of energy segments comprises:

sequentially determining the fluctuation differences between the pixels in each row in the target image according to a sequence of the pixels in each row of pixels in the target image; and

performing division when the fluctuation difference between two pixels in each row is greater than a preset fluctuation threshold, to obtain one energy segment.

3 . An electronic device, comprising:

a processor; and

a memory used for storing processor-executable instructions;

wherein, the processor is configured to implement, with low power and a small amount of calculation, the visual image processing method based on a strain mechanism according to claim 2 .

4 . The visual image processing method based on a strain mechanism according to claim 1 , wherein combining energy segments that have the same photon energy and are adjacent to each other according to the initial spatial positions, the end spatial positions and the lengths corresponding to the plurality of energy segments, to obtain a plurality of energy fields comprises:

determining whether energy segments with the same photon energy are adjacent to each other according to the initial spatial positions and the end spatial positions corresponding to the energy segments with the same photon energy; and

splicing energy segments that are adjacent to each other and have the same photon energy according to the lengths of the energy segments to obtain one energy field.

5 . An electronic device, comprising:

a processor; and

a memory used for storing processor-executable instructions;

wherein, the processor is configured to implement, with low power and a small amount of calculation, the visual image processing method based on a strain mechanism according to claim 4 .

6 . A non-transitory computer-readable storage medium, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to implement the visual image processing method based on a strain mechanism according to claim 4 .

7 . An electronic device, comprising:

a processor; and

a memory used for storing processor-executable instructions;

wherein, the processor is configured to implement, with low power and a small amount of calculation, the visual image processing method based on a strain mechanism according to claim 1 .

Assignments (1)

Priority Claims (1)

References Cited (15)