Camera stain detecting method and device

Abstract

The invention discloses a camera stain detecting method and device. The method comprises that in each detection period, brightness is sampled twice at a first sampling step t1 and once at a second sampling step t2, whether the brightness error of each sampled brightness value and the rate of brightness error change are respectively within corresponding confidence intervals is determined, if yes, the determination result is 1, and otherwise the determination result is 0; and logic AND operation is carried out on determination results of the three sampled brightness values, an operation result of the detection period is output, if the output operation result is 1 it is determined that no stain exists in pixels sampled in the detection period and a next detection period is moved to, and if the output operation result is 0 it is determined that a stain exists in the pixels sampled in the detection period and detection is stopped. The method and device can be used to effectively analyze stains in an image shot by the camera, and improve the detection accuracy.

Description

technical field [0001] The invention relates to the technical field of optical device testing, in particular to a method and device for detecting camera stains. Background technique [0002] The traditional method of camera image stain test is to judge whether there is stain in the image according to the way of brightness deviation. That is, a theoretical brightness value and a set of upper and lower deviation thresholds are set, and only pixels outside the brightness range are considered as stains. However, in the actual production test, there is a kind of stain, as shown in Table 1, which is the brightness value distribution table of the pixels in the picture taken by the stain camera. Assume that the theoretical brightness value of the image is 140, the upper and lower threshold values ​​are 15, the brightness value indicated by "*" in Table 1 is 150, and there is a piece of dirt with a brightness value of 130 in the middle area. In this case, the dirty and normal value...

AI Technical Summary

Problems solved by technology

In this case, the dirty and normal values ​​exist in the theoretical reasonable range at the same time. The traditional stain judgment method is used to think that the pixels in Table 1 have no stains, but in fact there are stains in the pixels in Table 1. Therefore, the traditional stain judgment method stain detection The accuracy rate is low and cannot meet the test needs

Images