Multistage network fusion round steel surface defect detection method and image acquisition device
A defect detection and surface image technology, applied in image analysis, image enhancement, image data processing, etc., can solve problems such as being easily affected by complex and changeable environments, insufficient algorithm generalization ability, and complex algorithm design process, etc., to achieve improved The effect of detection accuracy, improvement of training accuracy and detection accuracy, and reasonable sample division
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0061] Such as figure 1 and Figure 8 As shown, this embodiment provides a multi-level network fusion round steel surface defect detection method, the algorithm frame diagram of the multi-level network fusion round steel surface defect detection method is as follows Figure 8 As shown, the detection method includes:
[0062] S100. Obtain a real-time round steel surface image, establish a round steel surface defect data set with the round steel surface image with defects on the round steel surface, and establish a round steel surface defect-free image sample with the round steel surface image without defects on the round steel surface;
[0063] S200, using the round steel surface defect data set to create the round steel surface defect detection network model of the Efficient Net classification priority network and YOLOv5 network fusion, the model can classify and identify the round steel surface defect type and locate the round steel surface defect position;
[0064] S300. U...
Embodiment 2
[0086] Such as Figure 9 As shown, this embodiment provides a multi-level network fusion round steel surface defect detection method, the algorithm frame diagram of the multi-level network fusion round steel surface defect detection method is as follows Figure 8 As shown, the detection method includes:
[0087] S100. According to the requirements of the actual application scene, design an image acquisition device for the specifications and diameters of the round steel, use the acquisition device to acquire real-time round steel surface images, and establish the round steel surface images with defects on the round steel surface into a round steel surface defect data set , the round steel surface image without defects on the round steel surface is established as a non-defective image sample on the round steel surface;
[0088] S200. Using the round steel surface defect data set to create a round steel surface defect detection network model that combines the Efficient Net class...
Embodiment 3
[0092] As shown in Figure 10, this embodiment provides an image acquisition device suitable for round steel with specifications and diameters of 50-130mm, and the device is used to obtain the multi-level network fusion round steel surface defect detection method In the image of the round steel surface, the setup consists of:
[0093] The device includes a bracket on which four line-scan cameras are evenly distributed on the circumference, and the angle between each line-scan camera and the horizontal line is 45°, and the angle between the line-scan camera and the vertical line is 45°. °, the line scan camera has a GIGE digital interface, and the collected round steel surface image is transmitted to the server remotely through the network, and the server can store the round steel surface image. Specifically, four line scan cameras are distributed at intervals of 90° in the radial direction of the round steel. The setup of four line-scan cameras ensures that the circle of the r...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More