Study On Machine Vision Detection Of PVC Coating On Bodywork Chassis

The safety stability,driving comfort and running acceleration of voiture are important indicators of voiture performance.They are also the requirements for people to choose models.However,the road is getting bumpier and bumpier Because of long-term sun exposure and voiture driving.The chassis bodywork constantly produces collision with the ground,and is easy to damage its surface anti-corrosion layer,losing anti-corrosion performance.Therefore,in order to effectively increase the service life and safety performance of the voiture,it will be added 0.5cm to 1cm thick PVC spray adhesive sprayed on the bodywork chassis to enhance the sealing under the bodywork chassis.Under the condition of industrial production,due to its low working efficiency,traditional PVC spray adhesive is easily affected by the state of operators,and there are many non-standard situations,which can hardly meet the requirements of test standards for performance in modern industrial production.In order to rapidly realize the inspection items in industry and combine artificial intelligence detection,the machine vision detection can effectively meet the demand of efficient production.The visual examination of machine the bodywork carriage,PVC coating is an analysis of the need system to be tested.According to the needs,the soft and hard structure are designed.The pvc dilute concentration will be adjusted effectively by testing the result.And pvc coating will be improve spray by setting under the same parameters.To enhancing the tightness between the face of the bodywork and the mechanical interstices.Firstly,based on the system requirements,the general process of the system is designed.According to the total design process for detecting position requirements are studied,and on the basis of the completing software and hardware of system simulation.Secondly,the 5-megapixel CCD camera is determined for image acquisition because of the image detection type and precision.After the determination of the camera and lens,the lighting system is confirmed.By adjusting the brightness of the light source,light resolution determined the brightness of the image exposure range from 120 to 204 grayscale,It is setting 120 grayscale for the system brightness.Thirdly,the detection process of the system is analyzed and related algorithm process based on configuration parallelism is designed.And the image processing process studied in depth.The image pretreatment is 30%,the image binarization threshold is 50,and the algorithm such as image segmentation,elimination of two-fold repeated nodes and result output is designed.The system can learn and optimize by itself.The detection algorithm is installed in the detection software as a whole by constantly replenishing the number of samples to prepare the newly emerging detection area.Finally,in the test environment,machine vision test of PVC adhesive coating under voiture body is verified of PVC performance.In the test environment without intervention that the system is tested every day for 16 hours.It verifies the feasibility of the system and stores the data result image for later data query and pivot.