Near-surface pedestrian detection method based on deep learning for UAVs in low illumination environments

Wang, Congqing; Luo, Dehan; Liu, Yang; Xu, Bin; Zhou, Yongjun

doi:10.1117/1.oe.61.2.023103

Cited by 8 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Referring to the literature, 25 the detection is considered correct when the IoU is higher than 0.2. With the criteria for correct detection of cracks, five metrics were used to evaluate the performance of the model, which are recall, 72 precision, 61 average precision (AP), 35 break even point (BEP), 73 and inference time 74 . Among them, the first four evaluations all reflect the accuracy of the model in detecting cracks, i.e., the higher the evaluation is, the higher the accuracy is.…”

Section: Methodsmentioning

confidence: 99%

“…33 Currently, there are two ways that fusing multimodal images to detect objects. The first one is to fuse multiple different images to obtain a clearer and more comprehensive image, 34,35 then the objects can be detected in the fused image. This method is mainly used to visually enhance the saliency of the objects, and its loss function uses relevant algorithms for evaluating image quality, 36,37 such as image standard deviation.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Low saliency crack detection based on improved multimodal object detection network: an example of wind turbine blade inner surface

Gao

Dai

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

.Accurate identification of cracks is of great significance for maintaining the health of the equipment. However, the low saliency of cracks in some composite or metal surfaces affects the detection accuracy of object detection algorithms. For example, small cracks on the inner surface of wind turbine blade (WTB) may be similar in color to the substrate or face complex background textures. Taking WTB cracks as low saliency crack samples, we propose a multimodal object detection convolutional neural network that fuses infrared images with visible images to detect cracks more accurately. The proposed network contains the CenterNet network with an existing fast and efficient mid-level fusion structure. First, we optimized the fusion structure to make it more suitable for extracting crack features. To address the problem that severe background interference in multimodal images affects the detection performance, we add channel attention to the fusion structure and train the improved network using a stepwise training method to enhance the framework’s ability to filter background interference information. Finally, the effectiveness of the improvements was verified by ablation experiments and feature map analysis, and the phenomena of wrong detection, missed detection, and repeated detection were reduced. The evaluation results show that the proposed multimodal object detection network is able to detect the low saliency WTB cracks more effectively, and the improvement of the network also results in a 6.22% increase in average precision. In addition, this method can be extended to other materials or scenes to identify very inconspicuous objects, replacing manual inspection in more challenging defect detection tasks.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Low saliency crack detection based on improved multimodal object detection network: an example of wind turbine blade inner surface

Gao

Dai

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

show abstract

“…The experimental results on the dataset of INRIA showed that the pedestrian target detection algorithm with domain adaptation had less classification error. Wang et al [10] designed an algorithm using image fusion and deep learning to improve the performance of unmanned aerial vehicles for detecting pedestrians on the ground in lowillumination environments and verified the excellent performance of the algorithm through experiments.…”

Section: Introductionmentioning

confidence: 96%

Research on Automatic Detection Algorithm for Pedestrians on the Road Based on Image Processing Method

Zhang¹

2023

IJACSA

View full text Add to dashboard Cite

Accurate detection of pedestrian targets can effectively improve the performance level of intelligent transportation and surveillance projects. In order to effectively enhance the accuracy of detecting pedestrian targets on the road, this paper first introduced the traditional pedestrian target detection algorithm, proposed the faster recurrent convolutional neural network (RCNN) algorithm to detect pedestrian targets, and improved it to make good use of the convolutional features at different scales. Finally, support vector machine (SVM), traditional Faster RCNN, and optimized Faster RCNN algorithms were compared by simulation experiments. The results showed that the optimized Faster RCNN algorithm had higher detection accuracy and recall rate, obtained a more accurate target localization frame, and detected faster than SVM and traditional Faster RCNN algorithms; the traditional Faster RCNN algorithm had higher detection accuracy and target frame localization accuracy than the SVM algorithm.

show abstract

“…Jin et al [ 11 ] utilized one emerging method based on YOLOv3 in high-density pedestrians detection situations and achieved good results. To improve the near-surface detection performance of UAVs in low illumination environments, Wang et al [ 12 ] proposed a U-type generative adversarial network (GAN) to fuse visible and IR images to generate color fusion images. Then, a YOLOv3 model combined with transfer learning was trained using the fused images and achieved good results.…”

Section: Introductionmentioning

confidence: 99%

Research on Pedestrian Detection Model and Compression Technology for UAV Images

Liu

Wang

Liu

2022

Sensors

View full text Add to dashboard Cite

The large view angle and complex background of UAV images bring many difficulties to the detection of small pedestrian targets in images, which are easy to be detected incorrectly or missed. In addition, the object detection models based on deep learning are usually complex and the high computational resource consumption limits the application scenarios. For small pedestrian detection in UAV images, this paper proposes an improved YOLOv5 method to improve the detection ability of pedestrians by introducing a new small object feature detection layer in the feature fusion layer, and experiments show that the improved method can improve the average precision by 4.4%, which effectively improves the pedestrian detection effect. To address the problem of high computational resource consumption, the model is compressed using channel pruning technology to reduce the consumption of video memory and computing power in the inference process. Experiments show that the model can be compressed to 11.2 MB and the GFLOPs of the model are reduced by 11.9% compared with that before compression under the condition of constant inference accuracy, which is significant for the deployment and application of the model.

show abstract

Near-surface pedestrian detection method based on deep learning for UAVs in low illumination environments

Cited by 8 publications

References 0 publications

Low saliency crack detection based on improved multimodal object detection network: an example of wind turbine blade inner surface

Low saliency crack detection based on improved multimodal object detection network: an example of wind turbine blade inner surface

Research on Automatic Detection Algorithm for Pedestrians on the Road Based on Image Processing Method

Research on Pedestrian Detection Model and Compression Technology for UAV Images

Contact Info

Product

Resources

About