Two-dimensional Image Analysis And Processing Under Ambient Light Source Changes

In machine vision system,the quality of acquired image is the basic information source for analysis and decision making.Hence,the quality of the image directly determines the accuracy of the machine vision.When the target object is not illuminated by itself,it is necessary to illuminate the target object by the ambient light source so that the characteristics of the target object are clearer.And this method can simplify the algorithm to ensure the stability of the system detection.There can be more ideal processing results in the image processing stage,and it is beneficial to obtain higher measurement accuracy.If the brightness of the ambient light source is too strong,it will cause excessive exposure,resulting in the saturation will hide a lot of important information;if the brightness of the ambient light source is too weak,the acquired image will be dark,resulting in a low contrast of the image.In this paper,according to the strength of the external light source lighting,the image of the target scene is analyzed,and an effective illumination processing method is proposed,that is,adaptive illumination technology based on the projector-camera system.In addition to the influence of surface reflectivity,the surface orientation of the target object is also an important factor affecting the surface illumination intensity of the target object under the condition of ambient lighting.Therefore,adaptive illumination technology should be based on three-dimensional reconstruction,not only be based on simple two-dimensional image processing.This paper introduces the basic principle of surface orientation reconstruction of target object used by Fourier transform profilometry,deduces theoretical formula,and measures the experimental application range under the hardware platform conditions in this paper.According to the surface height of target object,the pixel matching between the projector and the camera is performed.Taking the specific experimental sample as an example,it is clarified that the accuracy of pixel matching directly determines the effectiveness and precision of the subsequent adaptive illumination technology.Then,combined with the working principle of the projector,adaptive illumination technology is proposed.The projector plays a key role in the adaptive illumination technology,compensating for surface reflectivity at each pixel to be inversely proportional to the brightness of the pixel under ambient light.This process is realized by an iterative algorithm between the projector and the camera,emphasizing different illumination intensities according to different surface orientations in the target scene.Under the two conditions that the ambient light sources are strong lighting and weak lighting,different iterations are performed to explain the effectiveness of the adaptive lighting technique.Finally,in the process of changing the grating pattern projected by the projector,it is found that the illumination compensation effect of the dual-frequency sinusoidal grating is seems to be better than the single frequency.Therefore,under different conditions of external ambient light and strong lighting,this paper explores that the effect of illumination compensation of how many frequency sinusoidal grating is the best.In order to avoid the affect of subjective factors,quality evaluation indicators are introduced to make an objective judgment from multiple angles.Then the best illumination compensation is chosen under different ambient lighting intensity to optimize the proposed adaptive illumination technology.