Systematic Study On Image Processing & Automatic Counting For The End Image Of Reinforcing Steel Bar

The counting of reinforcing steel bar is an important tache during producing and selling. At present, almost all reinforcing steel factories count bundled bar one by one by workers time after time, it is low efficiency, laborious, and may cause high error. Therefore, it is a crucial question for them to develop an automatic system for reinforcing steel bar counting in order to alleviate labour intensity and improve the efficiency and veracity of counting.In this paper, a method of bundled reinforcing steel bar auto-counting based on digital image processing is brought forward after analyzing and processing the end image of reinforcing steel bar. And a processing and auto-counting system of reinforcing steel bar end image is developed in Visual C++ language and counting of bounded reinforcing steel bar is achieved.1. A handy image collection and processing system of reinforcing steel bar is constructed. The vital factors which influent the image quality, just like illumination, time, pose of bundled reinforcing steel bar, are designed for the circumstance of the reinforcing steel bar.2. The common theories and methods of compute image processing are discussed in this paper. The image preprocessing methods, just like gradation processing, image enhancement, threshold segmentation, are compared and analyzed. At last, the preprocessing methods to end image of reinforcing steel bar are confirmed.3. A method of bounded reinforcing steel bar auto-counting based on digital image processing is brought forward. The techniques of preprocessing to reduce image noises, and segment ate image, and eliminate conglutination are studied. The characteristic parameters of end image of reinforcing steel bar are distilled by connected region labeling method.4. A processing and auto-counting system of reinforcing steel bar end image is developed, which involves five modules: image adjustment, image processing, recognition and counting, automatic counting, accessorial operation. It not only can demonstrate the processing and counting algorithm steply, but also can realize the automatic counting of reinforcing steel bar directly and quickly.5. Polygon lasso and rubber tools are designed, and image processing methods, like image enhancement, threshold segmentation, and morphology operation, are adopted in the system. Choosing of the objective image is achieved, and the process of counting predigested, and the image noises is reduced, and the veracity and efficiency of counting are improved at last.6. Boundaries tracking-area filled algorithm is ameliorated. And on the basis of analyzing the labeling conflict when labeling connected region ordinally, a method of connected region labeling based on rescaning is brought forward and the exactly labeling of connected region is preferably achieved.