Research On Defect Detection Method Of E-TPU Midsole Based On Machine Vision

Thermoplastic polyurethane elastomer foaming material(E-TPU)is a new type of environmental protection and high elastic material.Its various products are rising in the international market.The shoe insole is one of the most popular products.At present,in industrial production,midsole defect detection relies on manual detection,which is costly and cannot meet the needs of industrial online real-time detection.In order to realize online and real-time defect detection,this paper proposes an effective defect detection method for the special shape of midsole.A method of double edge detection for midsole product based on improved Otsu method is proposed.The midsole has a relatively special shape,and the flank of the product interferes with the surface detection to a great extent.Therefore,the extraction of the double edge has become an indispensable prework,which not only plays an important role in the detection of edge defects,but also needs to separate the surface and the flank to avoid mutual interference between the detection of flank and surface defects.This method adopts the improved Otsu method to process the midsole image in two steps.It respectively combines with the Weighted Object Variance method(WOV)and the Neighborhood Valley-Emphasis method(NVE)to calculate the optimal threshold.This method can effectively and clearly extract the double edges of the midsole.The precision rate is 95.58%,the average running time is 1.8s.The experiment demonstrates that the proposed method in this paper has good detection performance and good applicability.An automatic detection and defect classification method for E-TPU midsole surface defects based on machine vision was proposed.The surface defects of the midsole of E-TPU rely on manual detection,and the qualified standards are uneven.Therefore,this paper proposes an E-TPU midsole surface defect detection method based on machine vision to achieve automatic detection and defect classification.The proposed method is divided into three parts:Color defect detection,block defect detection,and linear defect detection.Color defect detection uses RGB three channel self-inspection to identify scorch and color pollution.Block defect detection uses superpixel segmentation and background prior mining to determine holes,impurities,and dirt.linear defect detection uses Gabor filter and Hough transform to detect indentation and convex marks.After Color defect detection,block defect detection and linear defect detection are simultaneously performed by parallel computing.The false positive rate(FPR₂)of the proposed method in this paper is 8.3%,the false negatives rate(FNR₂)of the hole is 4.7%,the FNR of indentation is 2.1%,and the running time does not exceed 1.6s.The test results show that this method can quickly and accurately detect various defects in the E-TPU midsole.