Exposure method and apparatus

Abstract

The invention provides an exposure method and apparatus. The method comprises the following steps: obtaining a global picture; calculating the global average gray value of the global picture; determining whether the global average gray value meets a preset global target gray scale condition or not; when the global average gray value does not meet the preset global target gray scale condition, adjusting the exposure magnitude of the camera; when the global average gray value meets the preset global target gray scale condition, detecting the object in the global picture and obtaining the local picture with the object; calculating the local average gray value of the local picture; determining whether the local average gray value meets a preset local target gray scale condition or not; and when the local average gray value does not meet the preset local target gray scale condition, adjusting the exposure magnitude of the camera. According to the invention, the self-adaptive exposure of a global picture and the self-adaptive exposure of a local object are integrated for use in a mutually assisting manner, which maximally satisfies the lighting condition for imaging required by the human face of a target object.

Description

technical field [0001] The invention relates to the field of exposure technology, in particular to an adaptive camera exposure method and device for fusing global and local objects in a picture. Background technique [0002] With the popularization of deep learning technology and massive data in machine vision, more and more camera-based machine vision products are becoming mature. However, robustness has always been a major criticism in the field of vision. In addition to the impact of deep neural network and massive training data on the stability of products based on visual inspection and recognition, the hardware itself also has a great impact on the recognition effect. Under complex lighting conditions, such as low illumination and backlighting, the quality of the pictures captured by the camera seriously restricts the accuracy of computer vision algorithms. [0003] The camera exposure size is mainly determined by the opening and closing of the aperture, the exposure ...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the present invention provides an exposure method and device to solve the problem that the existing camera exposure method has poor adaptability to changes in ambient light, and objects in pictures often appear overexposed or underexposed in low-light environments or under backlighting. the problem of the situation

Images