CN-121982365-A - ATE image detection method, device and equipment

CN121982365ACN 121982365 ACN121982365 ACN 121982365ACN-121982365-A

Abstract

The application is applicable to the technical field of semiconductor testing, and provides an ATE image detection method, device and equipment, wherein the method comprises the steps of acquiring a first image; dividing the first image into a plurality of grids, determining grid brightness gain values of each grid, wherein the grids comprise a plurality of pixel points, the grid brightness gain values are used for correcting the grid brightness values, determining the pixel brightness gain values of all the pixel points in the first image, the pixel brightness gain values are obtained according to the grid brightness gain values of surrounding grids of a target grid where the pixel points are located, correcting the pixel brightness values of all the pixel points in the first image to obtain a second image, and the pixel brightness values are corrected by adopting the pixel brightness gain values of the pixel points. By using the ATE image detection method provided by the disclosure, the visibility of extremely light dirt can be improved.

Inventors

ZHOU YIMING
LI HONG
CHEN JUNREN
ZHANG HAOZHE

Assignees

深圳市辰卓科技股份有限公司

Dates

Publication Date: 20260505
Application Date: 20251222

Claims (10)

1. A method of ATE image detection, the method comprising: Acquiring a first image; Dividing the first image into a plurality of grids, and determining a grid brightness gain value of each grid, wherein the grids comprise a plurality of pixel points; Determining pixel brightness gain values of all pixel points in the first image, wherein the pixel brightness gain values are obtained according to grid brightness gain values of surrounding grids of a target grid where the pixel points are located; and correcting the pixel brightness value of each pixel point in the first image to obtain a second image, wherein the pixel brightness value is corrected by adopting the pixel brightness gain value of the pixel point.
2. The method of claim 1, wherein determining a grid brightness gain value for each grid comprises: Determining grid brightness values of a plurality of grids in the first image; screening target grid brightness values from the grid brightness values of a plurality of grids; and obtaining a grid brightness gain value of the grid according to the grid brightness value of the grid and the target grid brightness value, wherein the larger the difference between the grid brightness value and the target grid brightness value is, the larger the grid brightness gain value is.
3. The method of claim 1, wherein determining pixel brightness gain values for each pixel point in the first image comprises: Obtaining a pixel brightness gain value of the pixel according to a grid brightness gain value of a surrounding grid of a target grid where the pixel is located and a target smoothing factor, wherein the target smoothing factor comprises at least one of a first smoothing factor and a second smoothing factor, the first smoothing factor is used for reducing the pixel brightness gain value, and the second smoothing factor is obtained according to a distance between a pixel position and a center point position of the first image.
4. The method of claim 1, wherein determining pixel brightness gain values for each pixel point in the first image comprises: and obtaining the pixel brightness gain value according to the relative position of the pixel point in the target grid, the grid brightness gain value of the target grid and the grid brightness gain values of surrounding grids of the target grid.
5. The method of claim 1, wherein determining a grid brightness gain value for each grid comprises: Adding a new grid at the periphery of the edge grid under the condition that the target grid is an edge grid in a plurality of grids, wherein the grid brightness gain value of the new grid is the same as the grid brightness gain value of the edge grid; Obtaining the pixel brightness gain value according to the relative position of the pixel point in the edge grid, the grid brightness gain value of the edge grid and the grid brightness gain values of the surrounding grids of the edge grid, wherein the surrounding grids of the edge grid comprise the newly added grid.
6. The method of claim 1, wherein the dividing the first image into a plurality of grids and determining a grid brightness gain value for each grid comprises: splitting the first image into sub-images of a plurality of color channels; Dividing each sub-image into a plurality of grids, and determining a grid brightness gain value of each grid in each sub-image; The correcting the pixel brightness value of each pixel point in the first image to obtain a second image comprises the following steps: Correcting pixel brightness values of all pixel points in all the sub-images to obtain all corrected sub-images; And synthesizing each corrected sub-image into the second image.
7. The method according to claim 1, wherein the method further comprises: and carrying out contrast enhancement processing on the second image to obtain a target image, wherein the target image is provided with a dirty contour.
8. The method of claim 2, wherein said determining grid brightness values for a plurality of said grids in said first image comprises: acquiring pixel brightness values of a plurality of pixel points in the grid; and taking the average value of the pixel brightness values of a plurality of pixel points in the grid as the grid brightness value of the grid.
9. An ATE image detection apparatus, comprising: an acquisition unit configured to acquire a first image; The dividing unit is used for dividing the first image into a plurality of grids and determining a grid brightness gain value of each grid, wherein the grids comprise a plurality of pixel points; The pixel brightness gain value unit is used for determining the pixel brightness gain value of each pixel point in the first image, wherein the pixel brightness gain value is obtained according to the grid brightness gain value of the surrounding grids of the target grid where the pixel point is positioned; the correction unit is used for correcting the pixel brightness value of each pixel point in the first image to obtain a second image, and the pixel brightness value is corrected by adopting the pixel brightness gain value of the pixel point.
10. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 8 when the computer program is executed.

Description

ATE image detection method, device and equipment Technical Field The application belongs to the technical field of semiconductor testing, and particularly relates to an ATE image detection method, an ATE image detection device and ATE image detection equipment. Background The camera module comprises an image chip and a lens, ATE (Automatic Test Equipment ) is key test equipment in the mass production process of the image chip, and the ATE equipment is used for providing an electrical environment so that the lens in the camera module can shoot images, and the shot images are ATE images and are images which are generated by the ATE equipment by impelling the lens to shoot. Because of the difference between the production process and the environment, the produced optical lenses are inevitably stained with dust or other tiny particles, and in order to ensure that the contaminated lenses cannot finally fall into the hands of consumers, certain testing steps can be carried out before leaving the factory to ensure that the contaminated lenses are screened out. The lens can be subjected to dirt detection by adopting a test technology (Particle On Class, POG), dirt contours often exist in ATE images shot by the dirt lens in a white field environment, and whether the dirt contours exist in the ATE images can be analyzed through a series of image processing and algorithm analysis, so that whether the lens is a dirt lens can be further distinguished. In an actual production environment, an ATE image photographed by a lens is inevitably affected by the lens to generate shadows, that is, visual features with high brightness, low periphery and high middle of the image appear in the ATE image, and the shadows with low periphery are lens shadows caused by the photographing of the lens. In such a scene, if there is extremely light dirt in the lens shadow, it is difficult to automatically grasp the extremely light dirt, whether by naked eyes or by a dirt algorithm, and further the extremely light dirt cannot be accurately identified. Disclosure of Invention The embodiment of the application provides an ATE image detection method, device and equipment, which can improve the visibility of dirt and facilitate the recognition of the dirt by a subsequent dirt algorithm. In a first aspect, an embodiment of the present application provides an ATE image detection method, where the method includes obtaining a first image, dividing the first image into a plurality of grids, determining a grid brightness gain value of each grid, where the grid brightness gain value includes a plurality of pixel points, correcting the grid brightness value, determining a pixel brightness gain value of each pixel point in the first image, where the pixel brightness gain value is obtained according to grid brightness gain values of surrounding grids of a target grid where the pixel point is located, correcting the pixel brightness value of each pixel point in the first image to obtain a second image, and where the pixel brightness value is corrected by using the pixel brightness gain value of the pixel point. In a possible implementation manner of the first aspect, the determining a grid brightness gain value of each grid includes determining grid brightness values of a plurality of grids in the first image, screening a target grid brightness value from the grid brightness values of the grids, obtaining the grid brightness gain value of the grid according to the grid brightness value of the grid and the target grid brightness value, and the larger the difference between the grid brightness value and the target grid brightness value is, the larger the grid brightness gain value is. In one possible implementation manner of the first aspect, the determining the pixel brightness gain value of each pixel point in the first image includes obtaining the pixel brightness gain value of the pixel point according to the grid brightness gain value of the surrounding grid of the target grid where the pixel point is located and a target smoothing factor, where the target smoothing factor includes at least one of a first smoothing factor and a second smoothing factor, the first smoothing factor is used for reducing the pixel brightness gain value, and the second smoothing factor is obtained according to a distance between a pixel point position and a center point position of the first image. In a possible implementation manner of the first aspect, the determining a pixel brightness gain value of each pixel point in the first image includes obtaining the pixel brightness gain value according to a relative position of the pixel point in the target grid, a grid brightness gain value of the target grid, and a grid brightness gain value of a surrounding grid of the target grid. In one possible implementation manner of the first aspect, the determining the grid brightness gain value of each grid includes adding a new grid around the edge grid if the target grid is an edge grid