Power transmission line construction machinery hidden danger detection method based on deep learning
A technology for construction machinery and transmission lines, which is applied in closed-circuit television systems, computer components, instruments, etc., can solve the problems of threatening the safe operation of transmission lines, manpower consumption, large vertical and horizontal spans of transmission lines, etc., and achieve high-precision channel hidden danger target detection , the effect of improving the accuracy rate
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0027] Such as Figure 1-2 As shown, the deep learning-based transmission line construction machinery hidden danger detection method of the present invention includes the following steps:
[0028] a. The camera captures pictures within the scope of the transmission line and transmits them to the server through the 4G network; specifically, the transmission through the 4G network to the server means that the camera includes access and control equipment for the 4G network, and the control equipment automatically transfers the camera to the server. The captured monitoring pictures are uploaded to the server, and the access device of the 4G network can be connected to the wireless network of the telecom operator for remote transmission of pictures.
[0029] b. Obtain and mark pictures containing hidden dangers of construction machinery on the server side, use the neural network model to train the marked pictures, and obtain the construction machinery monitoring model for detection...
Embodiment 2
[0032] On the basis of Example 1, in the transmission line construction machinery detection system in a certain province, the camera captured 44,000 on-site pictures in a certain period of time during the capture process of the transmission line area, including 18,596 pictures containing hidden dangers of construction machinery and With 25404 pictures that do not contain construction machinery, using the construction machinery detection method proposed by the present invention, the false positive rate of identification of hidden dangers of construction machinery is 5.2%, the false positive rate of identification is 7.2%, and the accuracy rate is 88.1%.
[0033] a. Use the neural network model proposed by the present invention to train the 18,596 labeled data sets containing hidden dangers of construction machinery, repeatedly perform data iterative training, and finally obtain the construction machinery detection model;
[0034] b. Obtain 2000 uploaded pictures newly captured b...
Embodiment 3
[0038] On the basis of Example 1, in the transmission line construction machinery detection system in a certain province, the camera captured 44,000 on-site pictures in a certain period of time during the capture process of the transmission line area.
[0039] Using steps similar to those in Example 1, using all the samples in the training set for repeated iterative training, in Example 2, the false negative rate of recognition was calculated to be 5.9%, the false positive rate of recognition to be 6.8%, and the recognition accuracy rate to be 85.5%, which met the technical requirements. .
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 

