Study On The Detection And Parameter Statistical Method Of Glass Fragments Under Impact Fragmentation

Safety glass compulsory product certification,enforced by our government,is a mandatory assessment product certification system on glass production,evaluating whether its safety performance accords with the national standards in order to protect the vast number of consumers' personal safety and health.To evaluate the possibility of the damage caused by the fragmentation,the fragmentation state test becomes an important part of the mechanical property inspection.Presently,there are more than 400 certified safety glass enterprises in China,each enterprise owing a safety glass product testing laboratory.But the test results is always determined personally by the naked eye.Since people are easily influenced by personality,mood,vision etc.,the manual inspection process is not only costly and time consuming,but also prone to errors,affecting the accuracy of the test results.Considering the above situation,a complete software for automatic detection system of glass debris is designed and developed in this paper,which is easily to get the glass fragment segmentation and the corresponding parameter statistics.In this paper,the criteria and the debris state test of safety glass are briefly described.By analyzing the histogram of glass fragment gray image and the characteristics of glass debris noise,the paper summarizes the characteristics and difficulties of the image processing application and glass debris detection,and a complex glass debris automatic detection application system has been put forward,which includes the image preprocessing,segmentation,statistics.In accordance with the quality inspection standards set by the glass manufacturer,this paper selects the appropriate camera,light source,computer and other hardware,and sets up a complete automatic inspection platform for the safety glass quality.An image processing algorithm based on background subtraction and a modified one shot scanning algorithm are developed to improve the detection speed.Based on Visual C++6.0,a complete set of automatic testing system for safety glass quality is developed,including the frame of the application program writing,testing and debugging,and the design and implementation of some software algorithms.Finally,this paper studies the design method of glass debris detection system and other related issues,in which a man-machine friendly debris detection system interface is designed and developed.After testing,the system meets the requirements of automatic detection of glass debris with the detection speed of 100 milliseconds per picture.This paper lays a good foundation for the further development of the quality inspection of the safety glass.