Smoke and fire detection and straw detection method and system based on target detection technology
A pyrotechnic detection and target detection technology, applied in the field of electronic information, can solve the problems of complex pyrotechnic image features, low recognition rate, and inability to judge and recognize
Pending Publication Date: 2022-08-02
东南数字经济发展研究院
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, in actual situations, the features of pyrotechnic images are very complex, and cannot be judged and identified by a single feature or a combination of several features.
The above traditional methods can be used in some specific scenarios, but they are severely restricted in natural scenarios, with low recognition rates and high false alarm rates
limit its wider application
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0031] The present invention provides a method for pyrotechnic detection and straw detection based on target detection technology, see figure 1 ,include:
[0032] Step 1): build a fireworks detection model and a straw detection model;
[0033] Step 2): based on the detection algorithm, under the collected data set, respectively train the fireworks detection model and the straw detection model;
[0034] Further, the detection algorithm is an optimized YOLOv3 algorithm.
[0035] Further, the optimization method of the YOLOv3 algorithm is specifically:
[0036] In category prediction, the Softmax loss function in the YOLOv3 algorithm is replaced with the Logistic function.
[0037] Specifically, the errors of the YOLOv3 algorithm mainly include confidence error, positioning error and classification error. In category prediction, YOLOv3 does not use the Softmax loss function, but uses the Logistic function instead, which better supports multi-label objects. classification task...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
According to the smoke and fire detection and straw detection method based on the target detection technology, a target detection algorithm is improved and optimized, the detection purpose is achieved, and finally the effectiveness of the method is verified through experiments; and then, an intersection is taken through a model fusion method for model fusion, a fused straw incineration model is obtained, and the effectiveness of the model is verified through experiments. On the basis of a straw incineration model, a straw incineration detection system scheme capable of being used in the actual process is constructed. The detection system is mainly divided into four modules, namely a front-end layer, a transmission layer, a background layer and a monitoring platform which are respectively responsible for acquiring front-end images, transmitting image data among the modules, performing data detection and algorithm analysis on transmitted images and displaying a final result on the monitoring platform. The straw detection system is applied to the whole actual process, and finally the purpose of detecting straw burning is achieved.
Description
technical field [0001] The invention relates to the technical field of electronic information, in particular, to a method and system for detecting fireworks and straws based on target detection technology. Background technique [0002] Straw incineration is harmful to all, and relevant environmental protection departments have formulated various protection policies for straw incineration. The current methods of judging straw incineration are mainly through manual work, sensors or remote sensing images, which consume a lot of manpower and material resources, and have many technical limitations. Considering that deep learning and fireworks detection and recognition are the main research hotspots at present, it is a good method to apply fireworks detection technology in straw burning. [0003] At present, there are many fireworks detection methods in life. In the indoor environment, we generally detect fireworks by physical (light measurement) and chemical (ionization) method...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06V20/52G06V10/82G06V10/80G06V10/764G06N3/04G06N3/08G08B17/12
CPCG06V20/52G06V10/82G06V10/80G06V10/764G06N3/08G08B17/125G06V2201/07G06N3/045Y02T10/40
Inventor 朱绳祖
Owner 东南数字经济发展研究院