Research On The Key Technology Of Image Pretreatment Container Number

Image preprocessing is to eliminate the irrelevant information in the original image,recover the useful real information,and enhance the detectability of related information.Image preprocessing is important for the recognition of subsequent image texture features and is crucial in the image processing process.This thesis uses research methods of literature analysis,comparison analysis and experiment simulation,taking the container image as research object,to explore the issues about container image preprocessing,proposing a bilateral filtering shadow removal method based on the container intrinsic image,improving the fuzzy restoration method of the container image motion blur parameter and the container image color enhancement method of Retinex,and demonstrating the advancement and feasibility of the proposed method through related experiments.The main work of the article is as follows:(1)Analysis of the formation mechanism of shadow blurring in container images.According to the shooting environment and formation process of container images,container characteristic analysis was conducted to analyze the shadow formation mechanism and fuzzy formation mechanism of container images.(2)Research on shadow elimination method for container images.By analyzing the shadow removal method based on the Poisson equation and the shadow removal method based on the shadow factor estimation,it is found that the detection of the shadow boundary is required.However,the shadow boundary of the container image is often detected inaccurately and will affect the shadow removal effect.This paper uses an image-based method.The shadow erasing method of the present image obtains the image 3-D proof image and restores the image illumination,thereby obtaining an unshaded container image and verifying it through experiments.(3)Research on the method of fuzzy image restoration for container images.Bispectral segmentation estimation method of image spectrogram is used to estimate the value of fuzzy angle.Improving the adaptive fuzzy length method is used to estimate the value of fuzzy length.According to the obtained motion blur parameters,the Wiener filtering method is used to obtain the fuzzy restored image,which was verificated by experiments.(4)Research on the method of color enhancement for container images.By analyzing several common methods of color image enhancement,this article proposes a color enhancement algorithm based on the improved Retinex for container images.Through the improved Retinex algorithm,the container images are enhanced and the image texture information becomes more obvious,which was verificated by experiments.(5)Research on the application of container image preprocessing.The process algorithm is used in the preprocessing process of container images,and the pre-processed images are identified by box number characters.Through the analysis of recognition rate,the feasibility of image preprocessing is proved.