Deep Learning-Based Bird’s Nest Detection on Transmission Lines Using UAV Imagery

Li, Jin; Yan, Daifu; Luan, Kuan; Li, Zeyu; Hong, Liang

doi:10.3390/app10186147

Cited by 40 publications

(24 citation statements)

References 22 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Research related to bird nest detection has also been carried out by J. Li, D. Yan, K. Luan, Z. Li, and H. Liang [12]. In this study, several architectures were used to detect bird nests, namely Faster RCNN, Faster RCNN with Focal Loss, Cascade RCNN, and ROI Mining Faster RCNN with mAP values of 78.29%, 78.99%, 79.19%, and 82.51%, respectively.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Object Detection to Identify Shapes of Swallow Nests Using a Deep Learning Algorithm

Indrajaya

Setiawan

Hartanto³

et al. 2022

khif

View full text Add to dashboard Cite

Object detection is basic research in the field of computer vision to detect objects in an image or video. the TensorFlow framework is a widely adopted framework to create object detection programs and models. In this study, an object detection program and model are designed to detect the shape of a swallow's nest which consists of three classes, namely oval, angular, and bowl. The purpose model creation is to find out the likeliness of the swallow's nest to the three classes for the swallow's nest sorting machine. The adopted architecture in the modeling is the MobileNet V2 FPNLite SSD since the model obtained from this architecture results in a good speed in detecting objects. Based on the evaluation results that has been carried out, the model can detect the shape of the swallow's nest which is divided into 3 classes, but in some cases swallow's nest are detected into two classes. This issues can still be handled by adjustmenting several parameterss to the object detection program. Results shows that the obtained mAP value of 61.91%, indicating the model can detect the shape of a swallow's nest moderately.

show abstract

Section: Discussionmentioning

confidence: 99%

“…In addition, some studies develop the MobileNet SSD architecture using the Feature Pyramid Network (FPN) module to improve detection accuracy [9]. Research on bird nest detection has been carried out several times, but in these studies, swallow nests were not used and bird nests were not divided into classes [10]- [12].…”

Section: Introductionmentioning

confidence: 99%

Object Detection to Identify Shapes of Swallow Nests Using a Deep Learning Algorithm

Indrajaya

Setiawan

Hartanto³

et al. 2022

khif

View full text Add to dashboard Cite

show abstract

“…Subsequently, based on Faster R-CNN [24], Wang et al [25] extracted proposal regions through different sliding windows and detected bird's nest targets. Li et al [26] proposed a deep learning-based framework for automatic bird nest detection-ROI mining and a faster RCNN. The detection accuracy of bird nest targets was improved by k-means clustering and the focal loss function.…”

Section: A Multi-stage Target Detection Algorithmmentioning

confidence: 99%

YOLOX++ for Transmission Line Abnormal Target Detection

Jing

Sun

et al. 2023

IEEE Access

View full text Add to dashboard Cite

The detection of abnormal targets in transmission lines plays a significant role in maintaining the stability and safety of transmission systems. To achieve improved detection performance for abnormal targets, we propose a new target detector based on YOLOX, called YOLOX++. First, a multiscale cross-stage partial network (MS-CSPNet) is designed, which fuses multiscale feature information and expands the receptive field of the target through channel combination. Furthermore, depthwise and dilated convolutions are introduced in an object decoupling head to better capture the long-range dependencies of objects in feature maps. Finally, the alpha loss function (𝛼-IoU) is introduced to optimize the localization of small objects. Experiments show that in a comparison with the YOLOX model, the YOLOX++ approach in this paper achieves 86.8% and 96.6% detection accuracies for high-voltage tower bird nest and power line insulator targets, respectively. On the PASCAL VOC dataset, the AP50 and APS are improved by 9.3% and 5.0% over those of YOLOX, respectively, showing that the YOLOX++ network possesses better robustness for small target detection.INDEX TERMS Target detection, transmission line anomaly target, small target detection, YOLOX.

show abstract

“…In such cases, a dataset can be created by taking many object photographs with a camera or by generating synthetic images with a game engine [25,26] according to the problem. CNN-based algorithms are frequently used to train DL networks for image classification [27][28][29]. There are some studies in the literature which make use of CNN-based image classification using pictures of real objects or synthetic objects [30].…”

Section: Related Workmentioning

confidence: 99%

Real UAV-Bird Image Classification Using CNN with a Synthetic Dataset

Öztürk

Erçelebi

2021

Applied Sciences

View full text Add to dashboard Cite

A large amount of training image data is required for solving image classification problems using deep learning (DL) networks. In this study, we aimed to train DL networks with synthetic images generated by using a game engine and determine the effects of the networks on performance when solving real-image classification problems. The study presents the results of using corner detection and nearest three-point selection (CDNTS) layers to classify bird and rotary-wing unmanned aerial vehicle (RW-UAV) images, provides a comprehensive comparison of two different experimental setups, and emphasizes the significant improvements in the performance in deep learning-based networks due to the inclusion of a CDNTS layer. Experiment 1 corresponds to training the commonly used deep learning-based networks with synthetic data and an image classification test on real data. Experiment 2 corresponds to training the CDNTS layer and commonly used deep learning-based networks with synthetic data and an image classification test on real data. In experiment 1, the best area under the curve (AUC) value for the image classification test accuracy was measured as 72%. In experiment 2, using the CDNTS layer, the AUC value for the image classification test accuracy was measured as 88.9%. A total of 432 different combinations of trainings were investigated in the experimental setups. The experiments were trained with various DL networks using four different optimizers by considering all combinations of batch size, learning rate, and dropout hyperparameters. The test accuracy AUC values for networks in experiment 1 ranged from 55% to 74%, whereas the test accuracy AUC values in experiment 2 networks with a CDNTS layer ranged from 76% to 89.9%. It was observed that the CDNTS layer has considerable effects on the image classification accuracy performance of deep learning-based networks. AUC, F-score, and test accuracy measures were used to validate the success of the networks.

show abstract

Deep Learning-Based Bird’s Nest Detection on Transmission Lines Using UAV Imagery

Cited by 40 publications

References 22 publications

Object Detection to Identify Shapes of Swallow Nests Using a Deep Learning Algorithm

Object Detection to Identify Shapes of Swallow Nests Using a Deep Learning Algorithm

YOLOX++ for Transmission Line Abnormal Target Detection

Real UAV-Bird Image Classification Using CNN with a Synthetic Dataset

Contact Info

Product

Resources

About