A remote judgment and repair method and system for smart city front-end faulty equipment
A technology for front-end equipment and faulty equipment, applied in the computer field, can solve problems such as failure to identify faulty equipment, inconsistent operation of front-end equipment, and difficulty in guaranteeing repair timeliness.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0049] refer to figure 1 , the present embodiment provides a remote judgment and repair method for front-end faulty equipment in a smart city, which includes the following steps:
[0050] S1. The server adopts a preset data crawling technology to obtain picture data from at least one urban network, and input the picture data into a preset front-end equipment prediction model for processing, so as to obtain the output of the front-end equipment prediction model Prediction results, and determine whether the prediction results are front-end equipment pictures; wherein, the prediction results include front-end equipment pictures or non-front-end equipment pictures; the front-end equipment prediction model is based on a preset deep convolutional neural network model and adopts Trained by supervised learning;
[0051] S2. If the prediction result is a picture of the front-end device, the server acquires the collection location and collection time of the picture data, and obtains th...
Embodiment 2
[0101] refer to figure 2 , this embodiment provides a remote judgment and repair system for front-end faulty equipment in a smart city for the remote judgment and repair method described in Embodiment 1, which includes:
[0102] The prediction result acquisition unit 10 is used to instruct the server to use a preset data crawling technology to obtain image data from at least one city network, and input the image data into a preset front-end equipment prediction model for processing, so as to obtain the The prediction result output by the front-end equipment prediction model, and determine whether the prediction result is a front-end equipment picture; wherein, the prediction result includes a front-end equipment picture or is a non-front-end equipment picture; the front-end equipment prediction model is based on a preset depth The convolutional neural network model is trained by supervised learning;
[0103] Designate the front-end equipment acquisition unit 20, which is use...
Embodiment 3
[0110] The only difference between this embodiment and Embodiment 2 is that the smart city front-end faulty equipment remote judgment and repair system in this embodiment also includes:
[0111] The manual marking unit is used to instruct the server to obtain a specified number of sample pictures collected in advance, and manually mark the sample pictures, so as to mark the front-end equipment in the sample pictures, so as to obtain the marked pictures;
[0112] A picture division unit, configured to instruct the server to divide the marked picture into a training picture and a verification picture according to a preset ratio;
[0113] The model training unit is used to instruct the server to call a preset deep convolutional neural network model, and input the training pictures into the deep convolutional neural network model for training to obtain a preliminary model;
[0114] A model verification unit, configured to instruct the server to perform verification processing on t...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More