An Improved Yolov5s based Real-time Spontaneous Combustion Point Detection Method

Zhou, Zhihao; Meng, YuWei; Yu, Rongdong; Yang, Tao; Ding, Lu; Wang, Zhan

doi:10.1109/cei52496.2021.9574458

Cited by 5 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CSP structure in YOLOv5s can reduce the calculation of the model and improve the detection speed of the model 52 . However, infrared targets often have few features, and the CSP structure cannot effectively extract the features of infrared targets.…”

Section: Csp_se Modelmentioning

confidence: 99%

Improved YOLOv5 infrared tank target detection method under ground background

Liang

Yan²,

Ren³

et al. 2023

Sci Rep

View full text Add to dashboard Cite

The detection precision of infrared seeker directly affects the guidance precision of infrared guidance system. To solve the problem of low target detection accuracy caused by the change of imaging scale, complex ground background and inconspicuous infrared target characteristics when infrared image seeker detects ground tank targets. In this paper, a You Only Look Once, Transform Head Squeeze-and-Excitation (YOLOv5s-THSE) model is proposed based on the YOLOv5s model. A multi-head attention mechanism is added to the backbone and neck of the network, and deeper target features are extracted using the multi-head attention mechanism. The Cross Stage Partial, Squeeze-and-Exclusion module is added to the neck of the network to suppress the complex background and make the model pay more attention to the target. A small object detection head is introduced into the head of the network, and the CIoU loss function is used in the model to improve the detection accuracy of small objects and obtain more stable training regression. Through these several improvement measures, the background of the infrared target is suppressed, and the detection ability of infrared tank targets is improved. Experiments on infrared tank target datasets show that our proposed model can effectively improve the detection performance of infrared tank targets under ground background compared with existing methods, such as YOLOv5s, YOLOv5s + SE, and YOLOV 5 s + Convective Block Attention Module.

show abstract

Section: Csp_se Modelmentioning

confidence: 99%

Improved YOLOv5 infrared tank target detection method under ground background

Liang

Yan²,

Ren³

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…The CSP structure in YOLOv5s can reduce the calculation of the model and improve the detection speed of the model 43 . However, few features of infrared dim targets are in the ground background, and the CSP structure cannot effectively extract the features of infrared dim targets.…”

Section: Csp_se Modelmentioning

confidence: 99%

Improved YOLOv5 infrared tank target detection method under ground background

Liang

Yan

Ren

et al. 2023

Preprint

View full text Add to dashboard Cite

To solve the problem of low target detection accuracy caused by the change of imaging scale, complex ground background and inconspicuous infrared target characteristics when infrared image seeker detects ground tank targets. In this paper, a You Only Look Once, Transform Head Squeeze-and-Excitation (YOLOv5s-THSE) model is proposed based on the YOLOv5s model, and a multihead attention mechanism is added to the backbone and neck of the network. The Cross Stage Partial, Squeeze-and-Exclusion (CSP_SE) module is added to the neck of the network, a small target detector is introduced into the head of the network, and the complete center section over union loss function is used in the model. Through various improvement measures, the background of the infrared target is suppressed, and the detection ability of the infrared tank target is improved. Experiments on infrared tank target data sets show that the model proposed in this paper can effectively improve the detection performance of infrared tank targets under ground background compared with several methods, such as YOLOv5s, YOLOv5s + SE, and YOLOV 5s + Convective Block Attention Module (CBAM).

show abstract

“…The Yolov5s network model is mainly composed of three parts: Backbone, Neck and Head, which are used to extract, enhance and predict the SCP images. The main part of the Yolov5s network is shown in Figure2: The Backbone layer is composed by Focus module, CBL (Conv+ Batch normalization+ Leakyrelu) module, SPP (Spatial Pyramid Pooling) feature pyramid, and Cross Stage Partial Network (CSPnet) [3] . The function of Focus is to slice the coal pile image to enhance the data information.…”

Section: 1construction Of the Recognition Model Of Scp Based On Yolov5smentioning

confidence: 99%

“…Therefore, translation and scaling are carried out during output, and the feature distribution to be learned by the original network is restored by introducing and training two reconstruction parameters. Because Leaky-relu has a small positive slope in the negative area [3] , it can carry out back propagation even for negative input values, which can solve the problem of neuron death caused by weight 0 in the image feature calculation of SCP.…”

Section: 1construction Of the Recognition Model Of Scp Based On Yolov5smentioning

confidence: 99%

“…These detectors are divided into two categories [2] . The first category uses smoke alarms, spectral sensors, and temperature sensing optical cables, this mode has a low false alarm rate and is suitable for a variety of applications, but its ability to locate the SCP is weak, and it needs to reach a certain level to detect the fire, which has a hysteresis; the second detection mode based on infrared cameras, the positioning of this mode is relatively accurate [3] . However, due to the presence of searchlights, coal pushers, bucket wheels and other unavoidable high temperature sources on the scene, the false alarm rate of this mode is high.…”

mentioning

confidence: 99%

See 1 more Smart Citation

A composite image recognition based approach for spontaneous combustion point detection and positioning in coal piles

Zhou

et al. 2022

J. Phys.: Conf. Ser.

Self Cite

View full text Add to dashboard Cite

The phenomenon of spontaneous combustion of coal piles often occurs in closed coal yards of power plants, automatic fire-fighting equipments are generally used to detect and locate the spontaneous combustion point. However, their performance (false alarm rate) are largely limited due to the complicated environment of the coal yard. To deal with this problem, this paper proposes a composite image recognition based method for detecting and positioning the spontaneous combustion point (SCP) in coal yard. Yolov5s is trained to efficiently detect SCPs and infrared information is fused to distinguish the realiability of SCP detection result via composite image registration method. In addition, an angle compensation algorithm is proposed to adjust the spray angle of the equipment for positioning the cooling area. The experimental results show that after integrating the above related technologies, the false alarm rate of the spontaneous combustion point in the scene is greatly reduced, and the work efficiency is effectively improved.

show abstract

An Improved Yolov5s based Real-time Spontaneous Combustion Point Detection Method

Cited by 5 publications

References 6 publications

Improved YOLOv5 infrared tank target detection method under ground background

Improved YOLOv5 infrared tank target detection method under ground background

Improved YOLOv5 infrared tank target detection method under ground background

A composite image recognition based approach for spontaneous combustion point detection and positioning in coal piles

Contact Info

Product

Resources

About