Detection of bird species related to transmission line faults based on lightweight convolutional neural network

Qiu, Zhibin; Zhu, Xuan; Liao, Chanjuan; Shi, Dazhai; Kuang, Yanjun; Li, Yanglin

doi:10.1049/gtd2.12333

Cited by 33 publications

(14 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…YOLO has several versions: v3, v4, and v5. The present work selects YOLOv3 for detecting objects from student pictures [ 22 , 23 ]. The basic structure of the YOLOv3 and the tiny YOLO are illuminated in Figures 5 and 6 , respectively.…”

Section: DL and Lightweight Object Detection Networkmentioning

confidence: 99%

See 1 more Smart Citation

Teachers’ Teaching Ability Promotion Strategies Based on Lightweight Deep Learning Combined with Target Detection Algorithm

Jin

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

With the popularization of standardized classrooms in colleges and universities, it is possible to collect video data of students’ class status through the camera device in the classroom. With abundant video data sources, it is easy to obtain big data of students’ class status images. Unstructured video big data is a topic worthy of research in improving teaching quality. First, the current teaching ability of teachers in colleges and universities is investigated, and its problems are found. Then, the You Only Look Once (YOLO) network in the object detection network is mainly studied. The deficiencies in the network structure are further explored and optimized. It is used in real classroom scenarios as well as on student expression detection problems. Finally, the proposed scheme is tested. The test results show that at present, 20% and 38% of teachers in higher vocational colleges think that they are “dissatisfied” with their classroom teaching and practical guidance ability. And 38% of teachers wanted to improve the bad situation. The accuracy of the proposed model for student expression detection is higher than that of faster-region convolutional neural network and mask-region convolutional neural network by more than 8%, higher than the YOLO v3 model by more than 4%, and higher than YOLO v3 Tiny model above 6%. The proposed model provides some ideas for the application of deep learning technology in the improvement of teachers’ teaching ability.

show abstract

Section: DL and Lightweight Object Detection Networkmentioning

confidence: 99%

“…YOLO has several versions: v3, v4, and v5. e present work selects YOLOv3 for detecting objects from student pictures [22,23]. e basic structure of the e Tiny YOLOv3 removes the last few layers of the MobileNetV3 and only retains all the previous blocks containing convolutional layers to replace the original DarkNet-53 network in YOLOv3.…”

Section: Yolo Cnnmentioning

confidence: 99%

Teachers’ Teaching Ability Promotion Strategies Based on Lightweight Deep Learning Combined with Target Detection Algorithm

Jin

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…The threshold of the complete intersection over union (CIoU) of non-maximum suppression (NMS) is 0.3, and the confidence level of the prediction box is 0.5. The calculation formula of cosine annealing learning rate updating with the number of iterations is shown in Equation (14).…”

Section: Settingsmentioning

confidence: 99%

“…The task of target detection (also be called target detection network in deep neural network,DNN) plays a significant role in the image processing system, which is to find out all the interested objects in the image and determine their positions and categories. Target detection network is the application of deep neural network in the field of computer vision and target detection, and many excellent application effects have been obtained [5][6][7][8][9][10][11][12][13][14]. Target detection network can be divided into one stage and two stages algorithms.…”

Section: Introductionmentioning

confidence: 99%

Multi-Objects Recognition and Self-Explosion Defect Detection Method for Insulators Based on Lightweight GhostNet-YOLOV4 Model Deployed Onboard UAV

Zhang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

In order to meet the application requirements on unmanned aerial vehicle (UAV) for daily inspection of insulators in overhead grid transmission lines, a lightweight GhostNet-YOLOV4 model is proposed to identify the insulators objects and detect their self-explosion defect meanwhile. In the proposed model, the lightweight GhostNet with full convolutional attention module (C-SE) model is embedded into the YOLOV4 as backbone feature extraction network, and the ordinary convolution and Mish activation function of enhanced feature extraction network in GhostNet-YOLOV4 are replaced by the depth separable convolution and ReLU6 activation function respectively. Meanwhile, the combination methods of data augmentation, label smoothing, K-Means algorithm, adaptive cosine annealing learning rate, and Focal Loss function are adopted to optimize the model. Finally, the class activation mapping (CAM) is used to visually analyze for the identification process of the proposed model. The example analysis results show that the object recognition accuracy mAP of the proposed model is 99.50% for insulator objects and self-explosion defect, and the image processing speed FPS is 2.53 frames/s. The proposed model meets the requirements of lightweight, detect accuracy and speed for UAV application.

show abstract

“…At present, the traditional manual inspection method has been gradually replaced by UAVs with greater flexibility and efficiency. In the transmission line inspection by UAV, the detection objects mainly include insulators [3][4][5], insulator self-explosion [6,7], vibration damper [8,9], bird species [10], and other components [11].…”

Section: Introductionmentioning

confidence: 99%

Transmission Line Object Detection Method Based on Label Adaptive Allocation

et al. 2022

View full text Add to dashboard Cite

Inspection of the integrality of components and connecting parts is an important task to maintain safe and stable operation of transmission lines. In view of the fact that the scale difference of the auxiliary component in a connecting part is large and the background environment of the object is complex, a one-stage object detection method based on the enhanced real feature information and the label adaptive allocation is proposed in this study. Based on the anchor-free detection algorithm FCOS, this method is optimized by expanding the real feature information of the adjacent feature layer fusion and the semantic information of the deep feature layer, as well as adaptively assigning the label through the idea of pixel-by-pixel detection. In addition, the grading ring image is sliced in original data to improve the proportion of bolts in the dataset, which can clear the appearance features of small objects and reduce the difficulty of detection. Experimental results show that this method can eliminate the background interference in the GT (ground truth) as much as possible in object detection process, and improve the detection accuracy for objects with a narrow shape and small size. The evaluation index AP (average precision) increased by 4.1%. Further improvement of detection accuracy lays a foundation for the realization of efficient real-time patrol inspection.

show abstract

Detection of bird species related to transmission line faults based on lightweight convolutional neural network

Cited by 33 publications

References 26 publications

Teachers’ Teaching Ability Promotion Strategies Based on Lightweight Deep Learning Combined with Target Detection Algorithm

Teachers’ Teaching Ability Promotion Strategies Based on Lightweight Deep Learning Combined with Target Detection Algorithm

Multi-Objects Recognition and Self-Explosion Defect Detection Method for Insulators Based on Lightweight GhostNet-YOLOV4 Model Deployed Onboard UAV

Transmission Line Object Detection Method Based on Label Adaptive Allocation

Contact Info

Product

Resources

About