The invention discloses a method, device, storage medium and terminal device for product defect detections and locations, wherein the method includes: according to the image data characteristics of products to be tested and the product defect complexity, selecting the neural network structure of detection models to be trained; obtaining training data, wherein the training data include training product pictures and standard categories and standard positions of product defects in the training product pictures; according to the training data, training the selected detection models of each neuralnetwork structure; selecting a best fitting detection model from the detection models after training to provide a detection server with the product defect detection, wherein the detection models are used for calculating according to the received product pictures to obtain predicted categories and predicted locations of the product defects in the received product pictures. Adopting the method, device, storage medium and terminal device for product defect detections and locations is capable of improving the accuracy of detecting the product defects.

Method, device, storage medium and terminal device for product defect detections and locations

The invention provides a product defect detection and location method comprising the following steps that the product image data to be detected are received, and a product defect detection request isgenerated; the product image data and the detection request are transmitted to the optimal server loading the detection model according to the load balancing and scheduling strategy; the prediction result outputted after prediction and calculation of the product image data by the detection module is received, wherein the prediction result includes the product defect position; and the correspondingresponse action is performed according to the prediction result outputted by the detection model. The product image acquired in real time is detected and judged and finally the position of the product defect and the category of the product defect can be acquired. Furthermore, iterative updating of the detection model is performed so that the detection model is enabled to be suitable for the latest requirements of the production environment, and the industrial production line can be significantly enhanced in the aspects of classification accuracy, expandability and standardization.

Product defect detection and location method, device and equipment and computer readable medium

The invention discloses a Faster R-CNN-based tire X-ray flaw detection method. According to the method, an effective tire X-ray flaw detection scheme which is suitable for all the tire industries andcan be popularized to other industrial detection fields is provided. According to the method, real suitable models are obtained through four stages such as data preparation, model training, parameteradjustment and model selection, so that the condition that most of existing tire X-ray flaw detection enterprises depend on a naked eye detection manner is greatly improved, the detection results aremore reliably, the detection processes are more rapid and the detection correctness is greatly improved. After being put into use, the method is a self-adaptive method, and along with the change of tire X-ray standards, the method is changed to achieve the effect that the method is also suitable under new standards.

Faster R-CNN-based tire X-ray flaw detection method

The invention provides a steel plate defect detecting method and apparatus, a device and a server. The steel plate detecting method includes the steps of receiving image data of a steel plate to be detected, and generating a defect detecting request according to the image data; monitoring operating states of a plurality of servers, and transmitting the image data of the steel plate to be detectedand the defect detecting request to a first server; receiving a prediction result from the first server, the prediction result being obtained by performing prediction calculation on the image data ofthe steel plate to be detected by a detecting model, the prediction result including detection qualified and defect results, and the defect result including a position and category of each defect of the steel plate to be detected; and performing a corresponding response action according to the prediction result. With the proposed method, the position and category of the defect can be obtained, andthe detection accuracy is improved.

Steel plate defect detecting method and apparatus, device and server

The invention discloses a tire X-ray defect detection method, comprising the following steps: S1, collecting normal X-ray images of all types of tires, cutting the images, and storing the images in adatabase. S2: Clustering all images. S3: Collect a large number of defective tire X-ray images, mark defective parts, defect types and tire types, and divide the images according to the clustering results. S4: For the classified defective tire X-ray images, preprocessing the images, then dividing the training set, test set and verification set, and building and training the target detection model.5, selecting a model according to that tyre type when the tyre image is tested, preprocessing the image and detecting with the model to obtain a detection result. This method can greatly improve theaccuracy of defect region detection.

A method for detect tire X-ray defects

The invention belongs to the technical field of image recognition and detection and specifically discloses a tire X-ray defect/flaw detection method based on the deep convolutional neural network The detection method comprises the steps of image preprocessing, data cleaning, convolutional neural network designing and training, parameter adjusting, input image preprocessing, model training and model testing According to the invention, tire x-ray images are classified based on the convolutional neural network Compared with the traditional method, image features can be automatically learned during the training process of the convolutional neural network, so that the defect of manually selecting features can be avoided The occurrence rate of accidental errors is reduced The detection result of the method is high in reliability and tire defect/flaw can be accurately detected The outstanding feature of applying the convolutional neural network to the image classification is that, a lot of time needs to be consumed during the training process of the model, while only a short period of time is required for the usage of the convolutional neural network model Therefore, the cost is saved The method is convenient for wide popularization

Li Quansheng

Papers

Tire X-ray defect/flaw detection method based on deep convolutional neural network