Identification method for fracture fault images of swing bolsters of railway wagon
A railway freight car and image recognition technology, applied in image enhancement, image analysis, image data processing and other directions, can solve the problems of low detection accuracy and poor stability of railway freight car bolsters, reduce the impact of category imbalance, improve Accuracy, improved robustness and precision
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0024] Specific implementation mode one: combine figure 1 , Figure 5 Describe this embodiment, the specific process of the image recognition method for the broken fault image of the railway freight car bolster in this embodiment is:
[0025] Step 1. Create a sample data set;
[0026] Step 2. Preliminary positioning of the area of the bolster component;
[0027] Step 3. Carry out self-adaptive contrast enhancement to the regional image of the initially intercepted bolster parts, so that the brightness and darkness of the initially intercepted regional images of the bolster parts are the same;
[0028] Since the angular distance of the imaging equipment at each site is different, the brightness and darkness of the collected images are different. Some images are too dark to clearly observe the fracture area of the bolster. Therefore, before entering the deep learning network, the image is adaptive to improve the contrast.
[0029] Step 4. Calculate the weight of the samp...
specific Embodiment approach 2
[0033] Embodiment 2: The difference between this embodiment and Embodiment 1 is that the sample data set is established in the step 1; the specific process is:
[0034] Imaging equipment is built on both sides of the railway track. After the truck passes through the equipment, a high-definition grayscale image is obtained; the image is a clear grayscale image. As truck components may be affected by rain, mud stains, oil stains, black paint and other natural or man-made conditions. Also, there may be differences in the images taken by different sites. Therefore, images of bolster parts vary widely. Therefore, in the process of collecting bolster image data, it is necessary to ensure diversity and try to collect all bolster images under various conditions.
[0035] In different types of bogies, the morphology of the bolster components varies. However, collection of bolster components for some less common bogie types is more difficult due to the wide variation in frequency bet...
specific Embodiment approach 3
[0042] Embodiment 3: The difference between this embodiment and Embodiment 1 or 2 is that in the step 2, the area of the bolster component is initially positioned; the specific process is:
[0043] According to prior knowledge such as hardware equipment, wheelbase information and related positions, the area of the bolster part is preliminarily intercepted from the image of the side camera.
[0044] Other steps and parameters are the same as those in Embodiment 1 or Embodiment 2.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


