Research And Implementation Of Key Technologies Of Cloth Fabric Defect Detection Platform Based On MachineVision

In the process of industrial production,quality control is an effective way to improve the core competitiveness of enterprises.In recent years,machine vision has been widely used in the field of industrial product quality control because of its unique advantages,such as non-contact,anti-interference,high speed,high precision and so on.Fabric defect detection is the main part of fabric quality control.At present,the detection of fabric defects in China is still dominated by traditional manual detection.In order to overcome the disadvantages of low efficiency,the false detection rate is high,and the leakage detection rate is high in the process of detection of man-made fabric defects.In order to solve the above situation and fabric defect detection based on machine vision has the characteristics of real-time and accuracy,the key technologies related to fabric defect detection in machine vision are studied and implemented.First of all,in order to solve the problem of fabric flaw detection,the image preprocessing methods such as image de-noising,image enhancement and image sharpening in image processing are used in the fabric detection system.The application of related algorithms is analyzed,and the above image preprocessing technology is used.In turn,the fabric images with better results and more clearly visible borders were obtained.Secondly,because the traditional Canny operator extracts the clear outline of the object,it is necessary to select its parameters manually and does not have self-adaptability.In this paper,an adaptive Canny edge detection method is proposed,which uses 3*3 neighborhood instead of 2*2 neighborhood in the Canny algorithm to calculate the gradient amplitude.Then,the Otsu algorithm is applied to the Canny threshold,so that the traditional Canny algorithm automatically obtains the threshold.The edge of the image is obtained by means of adaptive threshold,which makes the detected edge more continuous and reduces the existence of false edge.Then,using the relevant theory of neural network and combining the particularity of fabric defects,a cloth detection method based on neural network is designed.This method uses the characteristics of BP neural network.Then the fabric is identified by the detection and learning function of the BP neural network.The experimental results show that the method can recognize the fabric defect.Finally,the software system is designed and implemented.multi-thread synchronization mod was used to deal with the video images shows and cloth fabric defect detection.The detection rate,false detection rate and leakage detection rate of the system are introduced.Experiments show that this algorithm is effective for cloth detection.