Small-scale moving target detection in aerial image by deep inverse reinforcement learning

Sun, Wei; Yan, Dashuai; Huang, Jie; Sun, Changhao

doi:10.1007/s00500-019-04404-6

Cited by 16 publications

(15 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar to the FRCN algorithm, the multi-scale CNN consists of two sub-networks, one of which is the AODN (Accurate Object Detection Network) and the other is the MS-OPN (Multi-scale Object Proposal Network). Firstly, in the design process of network structure, the integration module is used to realize the depth feature extraction of remote sensing images with multi-scale information [25,26]. However, the extracted features are inconsistent with each other due to the change of target size.…”

Section: Multi-scale Change-oriented Target Detection Methods For Remote Sensing Imagesmentioning

confidence: 99%

Application of local fully Convolutional Neural Network combined with YOLO v5 algorithm in small target detection of remote sensing image

Liu

et al. 2021

PLoS ONE

163

View full text Add to dashboard Cite

This exploration primarily aims to jointly apply the local FCN (fully convolution neural network) and YOLO-v5 (You Only Look Once-v5) to the detection of small targets in remote sensing images. Firstly, the application effects of R-CNN (Region-Convolutional Neural Network), FRCN (Fast Region-Convolutional Neural Network), and R-FCN (Region-Based-Fully Convolutional Network) in image feature extraction are analyzed after introducing the relevant region proposal network. Secondly, YOLO-v5 algorithm is established on the basis of YOLO algorithm. Besides, the multi-scale anchor mechanism of Faster R-CNN is utilized to improve the detection ability of YOLO-v5 algorithm for small targets in the image in the process of image detection, and realize the high adaptability of YOLO-v5 algorithm to different sizes of images. Finally, the proposed detection method YOLO-v5 algorithm + R-FCN is compared with other algorithms in NWPU VHR-10 data set and Vaihingen data set. The experimental results show that the YOLO-v5 + R-FCN detection method has the optimal detection ability among many algorithms, especially for small targets in remote sensing images such as tennis courts, vehicles, and storage tanks. Moreover, the YOLO-v5 + R-FCN detection method can achieve high recall rates for different types of small targets. Furthermore, due to the deeper network architecture, the YOL v5 + R-FCN detection method has a stronger ability to extract the characteristics of image targets in the detection of remote sensing images. Meanwhile, it can achieve more accurate feature recognition and detection performance for the densely arranged target images in remote sensing images. This research can provide reference for the application of remote sensing technology in China, and promote the application of satellites for target detection tasks in related fields.

show abstract

Section: Multi-scale Change-oriented Target Detection Methods For Remote Sensing Imagesmentioning

confidence: 99%

Application of local fully Convolutional Neural Network combined with YOLO v5 algorithm in small target detection of remote sensing image

Liu

et al. 2021

PLoS ONE

163

View full text Add to dashboard Cite

show abstract

“…It can adjust the changes of the image under different illuminations through the logarithmic conversion of the pixel value. In addition, rendering is also a kind of image processing, and reasonable rendering can make the image clearer [15,16]. Recently, Finlayson proposed a new intrinsic image algorithm, and by calibrating the camera's sensitive equipment, the color image is converted into one-dimensional gray-scale image, thus eliminating the influence of shadow.…”

Section: Eigen Imagementioning

confidence: 99%

Detection and Adaptive Video Processing of Hyperopia Scene in Sports Video

Qing-jie

Dong

2021

Complexity

View full text Add to dashboard Cite

In the research of motion video, the existing target detection methods are susceptible to changes in the motion video scene and cannot accurately detect the motion state of the target. Moving target detection technology is an important branch of computer vision technology. Its function is to implement real-time monitoring, real-time video capture, and detection of objects in the target area and store information that users are interested in as an important basis for exercise. This article focuses on how to efficiently perform motion detection on real-time video. By introducing the mathematical model of image processing, the traditional motion detection algorithm is improved and the improved motion detection algorithm is implemented in the system. This article combines the advantages of the widely used frame difference method, target detection algorithm, and background difference method and introduces the moving object detection method combining these two algorithms. When using Gaussian mixture model for modeling, improve the parts with differences, and keep the unmatched Gaussian distribution so that the modeling effect is similar to the actual background; the binary image is obtained through the difference between frames and the threshold, and the motion change domain is extracted through mathematical morphological filtering, and finally, the moving target is detected. The experiment proved the following: when there are more motion states, the recall rate is slightly better than that of the VIBE algorithm. It decreased about 0.05 or so, but the relative accuracy rate increased by about 0.12, and the increase ratio is significantly higher than the decrease ratio. Departments need to adopt effective target extraction methods. In order to improve the accuracy of moving target detection, this paper studies the method of background model establishment and target extraction and proposes its own improvement.

show abstract

“…In terms of relative size, if the target occupies 0.1 times the size of the whole picture, it can be considered as a small target [4][5][6]. The traditional target detection algorithm is easy to cause misdetection and missed detection of small targets in these image processing, and the detection rate of small targets is low, so small target detection is the focus and difficulty in this field [7][8][9].…”

Section: Introductionmentioning

confidence: 99%

Small Target Detection Algorithm Based on Transfer Learning and Deep Separable Network

Wang

et al. 2021

Journal of Sensors

View full text Add to dashboard Cite

With the development of deep learning, target detection from vision sensor has achieved high accuracy and efficiency. However, small target detection remains a challenge due to inadequate use of semantic information and detailed texture information of underlying features. To solve the above problems, this paper proposes a small target detection algorithm based on Mask R-CNN model which integrates transfer learning and deep separable network. Firstly, the feature pyramid fusion structure is introduced to enhance the learning effect of low-level and high-level features, especially to strengthen the information channel of low-level feature and meanwhile optimize the feature information of small target. Secondly, the ELU function is used as the activation function to solve the problem that the original activation function disappears in the negative half axis gradient. Finally, a new loss function F-Softmax combined with Focal Loss was adopted to solve the imbalance of positive and negative sample proportions. In this paper, self-made data set is used to carry out experiments, and the experimental results show that the proposed algorithm makes the detection accuracy of small targets reach 66.5%.

show abstract

Small-scale moving target detection in aerial image by deep inverse reinforcement learning

Cited by 16 publications

References 29 publications

Application of local fully Convolutional Neural Network combined with YOLO v5 algorithm in small target detection of remote sensing image

Application of local fully Convolutional Neural Network combined with YOLO v5 algorithm in small target detection of remote sensing image

Detection and Adaptive Video Processing of Hyperopia Scene in Sports Video

Small Target Detection Algorithm Based on Transfer Learning and Deep Separable Network

Contact Info

Product

Resources

About