A Human–Computer Fusion Framework for Aircraft Recognition in Remote Sensing Images

Li, Xiaobin; Jiang, Bitao; Wang, Shengjin; Shen, Li; Fu, Yan

doi:10.1109/lgrs.2019.2918955

Cited by 4 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…e calculation of F1-Score is shown in Journal of Electrical and Computer Engineering [35] Figure 5 in [35] 73. 16 33.5 UAV-YOLO [36] Figure 1 in [36] 74.68 30.12 RFN [37] ResNet-101 79.1 6.5 SigNMS [38] VGG-16 80.6 6.7 Improved-YOLOv3 [39] Figure 4 in [39] 86.42 25.8 MRFF-YOLO [40] Figure 5 in [40] 87. 16…”

Section: Resultsmentioning

confidence: 99%

“…16 33.5 UAV-YOLO [36] Figure 1 in [36] 74.68 30.12 RFN [37] ResNet-101 79.1 6.5 SigNMS [38] VGG-16 80.6 6.7 Improved-YOLOv3 [39] Figure 4 in [39] 86.42 25.8 MRFF-YOLO [40] Figure 5 in [40] 87. 16…”

Section: Resultsmentioning

confidence: 99%

“…A target detection model based on Faster R-CNN was proposed, which could detect aircraft effectively, obtaining better performance than mature target detection networks. Li et al [16] aimed at the problem of recognizing aircraft in remote sensing images that contained multiple objects and background; a human-computer fusion framework that combined the advantages of human and computer was proposed. Wu et al [17] aimed at the problem that aircraft targets were usually small and the cost of manual annotation was very high; a simple yet efficient aircraft detection algorithm called Weakly Supervised Learning in AlexNet (AlexNet-WSL) was proposed to know detectors with only image-level annotations.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Aircraft Detection for Remote Sensing Images Based on Deep Convolutional Neural Networks

Zhou

Yan

Shan

et al. 2021

Journal of Electrical and Computer Engineering

View full text Add to dashboard Cite

Aircraft detection for remote sensing images, as one of the fields of computer vision, is one of the significant tasks of image processing based on deep learning. Recently, many high-performance algorithms for aircraft detection have been developed and applied in different scenarios. However, the proposed algorithms still have a series of problems; for instance, the algorithms will miss some small-scale aircrafts when applied to the remote sensing image. There are two main reasons for the problem; one reason is that the aircrafts in the remote sensing image are usually small in size, leading to detecting difficulty. The other reason is that the background of the remote sensing image is usually complex, so the algorithms applied to the scenario are easy to be affected by the background. To address the problem of small size, this paper proposes the Multiscale Detection Network (MSDN) which introduces a multiscale detection architecture to detect small-scale aircrafts. With the intention to resist the background noise, this paper proposes the Deeper and Wider Module (DAWM) which increases the perceptual field of the network to alleviate the affection. Besides, to address the two problems simultaneously, this paper introduces the DAWM into the MSDN and names the novel network structure as Multiscale Refined Detection Network (MSRDN). The experimental results show that the MSRDN method has detected the small-scale aircrafts that other algorithms missed and the performance indicators have higher performance than other algorithms.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Aircraft Detection for Remote Sensing Images Based on Deep Convolutional Neural Networks

Zhou

Yan

Shan

et al. 2021

Journal of Electrical and Computer Engineering

View full text Add to dashboard Cite

show abstract

“…Recall [102] Sensitivity [101] True Positive Rate (TPR) [101] Overall Accuracy [103] Detection Probability [68] Hit Rate [104] TP TP+FN PA for positives Precision [102] Positive Predictive Value (PPV) [101] TP TP+FP UA for positives Specificity [105] True Negative Rate (TNR) [101] TN TN+FP PA for negatives Negative Predictive Value (NPV) [101] TN TN+FN UA for negatives False Positive Rate (FPR) [106] Probability of False Detection [107] False Alarm Probability [100] FP TN+FP 1− (PA for negatives) False Negative Rate (FNR) 1 Missing Detection Probability [100] Missing Alarm [108] Misidentification Score [109] FN TP+FN 1− (PA for positives) False Discovery Rate (FDR) 1 False Alarm Probability [68] Commission Error [110] FP TP+FP 1− (UA for positives) Balanced Accuracy [101] Intersection-over-Union (IoU) [99] Jaccard Index [115] Figure 3 above summarizes the frequency at which each accuracy measure is used by papers that focus on binary and multiclass classification types, as well as by scene classification, object detection, semantic segmentation, and instance segmentation applications. A comparison of the graphs indicates that some measures (for example, precision and recall) are used for all types of classification applications, although it is notable that no single measure is used by every single study, even within one category of applications (e.g., multiclass scene identification).…”

Section: Overall Accuracymentioning

confidence: 99%

“…This is particularly important for metrics that are given general names such as average accuracy or accuracy. Lack of consistency in the literature as to what is meant by some measures is not just limited to F1, AP, and mAP, but includes other terms, such as false alarm rate/probability (e.g., compare [100,103]), reinforcing the importance of this issue.…”

Section: Clarity In Terminologymentioning

confidence: 99%

Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies—Part 1: Literature Review

2021

View full text Add to dashboard Cite

Convolutional neural network (CNN)-based deep learning (DL) is a powerful, recently developed image classification approach. With origins in the computer vision and image processing communities, the accuracy assessment methods developed for CNN-based DL use a wide range of metrics that may be unfamiliar to the remote sensing (RS) community. To explore the differences between traditional RS and DL RS methods, we surveyed a random selection of 100 papers from the RS DL literature. The results show that RS DL studies have largely abandoned traditional RS accuracy assessment terminology, though some of the accuracy measures typically used in DL papers, most notably precision and recall, have direct equivalents in traditional RS terminology. Some of the DL accuracy terms have multiple names, or are equivalent to another measure. In our sample, DL studies only rarely reported a complete confusion matrix, and when they did so, it was even more rare that the confusion matrix estimated population properties. On the other hand, some DL studies are increasingly paying attention to the role of class prevalence in designing accuracy assessment approaches. DL studies that evaluate the decision boundary threshold over a range of values tend to use the precision-recall (P-R) curve, the associated area under the curve (AUC) measures of average precision (AP) and mean average precision (mAP), rather than the traditional receiver operating characteristic (ROC) curve and its AUC. DL studies are also notable for testing the generalization of their models on entirely new datasets, including data from new areas, new acquisition times, or even new sensors.

show abstract

Feature Balance for Fine-Grained Object Classification in Aerial Images

Zhao

Tong

Yao

et al. 2022

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Object counting is a hot topic in computer vision, which aims to estimate the number of objects in a given image. However, most methods only count objects of a single category for an image, which cannot be applied to scenes that need to count objects with multiple categories simultaneously, especially in aerial scenes. To this end, this paper introduces a Multi-category Object Counting (MOC) task to estimate the numbers of different objects (cars, buildings, ships, etc.) in an aerial image. Considering the absence of a dataset for this task, a large-scale Dataset (NWPU-MOC) is collected, consisting of 3,416 scenes with a resolution of 1024 × 1024 pixels, and well-annotated using 14 fine-grained object categories. Besides, each scene contains RGB and Near Infrared (NIR) images, of which the NIR spectrum can provide richer characterization information compared with only the RGB spectrum. Based on NWPU-MOC, the paper presents a multi-spectrum, multicategory object counting framework, which employs a dualattention module to fuse the features of RGB and NIR and subsequently regress multi-channel density maps corresponding to each object category. In addition, to modeling the dependency between different channels in the density map with each object category, a spatial contrast loss is designed as a penalty for overlapping predictions at the same spatial position. Experimental results demonstrate that the proposed method achieves state-ofthe-art performance compared with some mainstream counting algorithms. The dataset, code and models are publicly available at https://github.com/lyongo/NWPU-MOC.

show abstract

A Human–Computer Fusion Framework for Aircraft Recognition in Remote Sensing Images

Cited by 4 publications

References 20 publications

Aircraft Detection for Remote Sensing Images Based on Deep Convolutional Neural Networks

Aircraft Detection for Remote Sensing Images Based on Deep Convolutional Neural Networks

Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies—Part 1: Literature Review

Feature Balance for Fine-Grained Object Classification in Aerial Images

Contact Info

Product

Resources

About