Performance evaluation of automatic object detection with post-processing schemes under enhanced measures in wide-area aerial imagery

Gao, Xin

doi:10.1007/s11042-020-09201-0

Cited by 6 publications

(19 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A filter is a software procedure that has the ability to modify the pixel values of images. These values relate to the color, appearance, brightness, and contrast of the image (Gao, 2020). Some methodologies combine two or more techniques to produce a hybrid method, such as R08, R14, and R15.…”

Section: Comparative Studymentioning

confidence: 99%

“…They are so massive and complex that it is a difficult task to process using traditional methods (Alahakoon et al, 2020). For example, the challenges of surveillance cameras are capturing and processing streaming images in a real-time environment, detecting and tracking objects in these images, storing information, and sending pre-alarms (Gao, 2020).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A comparative study of intelligent surveillance with big data: methods, strengths, weaknesses, and future directions

Habeeb

Abdulkhudhur

Habeeb

2022

KJS publishes peer-review articles in Mathematics, Computer Sci

View full text Add to dashboard Cite

Various challenges in intelligent surveillance with big data have attracted researchers to propose and design methods and techniques to solve these challenges. These methods have become an important part of healthcare, education, travel, security, and geographic research in our cities. They can monitor vehicles, count people, predict the intensity of tropical cyclones and give advance warnings of erratic behaviors or actions. However, the expansion of complex scenarios and big video data produces many papers describing how this data can be processed, analyzed, saved, and retrieved faster. The progress of these new trends is insufficient for intelligent surveillance solutions with big data. Hence, this research covers the academic literature with a comparative study of intelligent surveillance methods with big data. The study identified 98 papers for review from 2017 to 2022, then only 22 of them were finally selected for comparison according to a set of inclusion criteria. The result obtained includes all major aspects of these methods, such as their methodologies, purposes, strengths, and future directions, over the past six years of research. The objective of this paper was achieved by practically investigating recent trends in intelligent surveillance and its sub-topics in our life, using big data analytics. I hope that this comparative study will be a challenge for researchers and graduate students to find the gaps faster and overcome the drawbacks of the current intelligent surveillance methods using complex scenarios.

show abstract

Section: Comparative Studymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A comparative study of intelligent surveillance with big data: methods, strengths, weaknesses, and future directions

Habeeb

Abdulkhudhur

Habeeb

2022

KJS publishes peer-review articles in Mathematics, Computer Sci

View full text Add to dashboard Cite

show abstract

“…The highlevel features provide semantically meaningful activities, though they could have a higher error rate in classification tasks. With the development of the convolutional neural net (CNN)-based computer vision applications, the accuracy of image classification, object detection, and image tracking has achieved better performance compared to the traditional methods like the post-processing method proposed by Gao et al [29,30]. This fact inspires many researchers to use CNNs to extract features [22,31].…”

Section: Related Workmentioning

confidence: 99%

An explainable and efficient deep learning framework for video anomaly detection

Shao

Tunc

et al. 2021

Cluster Comput

View full text Add to dashboard Cite

“…We also adapted two saliency enhancement algorithms from the multiscale morphological analysis (MMA) [12] and the curvelet-Duda [31] algorithm (Duda). We implemented two versions on each of the nine algorithms: (1) using our prior binarization methods [27], [45] for simple thresholding, and (2) replacing the binarization step with the proposed spatial-processing or spatio-temporal processing. We refer to these two approaches as "before" and "after" the proposed post-processing in our experimental analysis, respectively.…”

Section: A Algorithms Tested and Experimental Setupmentioning

confidence: 99%

“…As future work, we plan to consider a parallel design for spatio-temporal processing via a high-performance computer [34] for better utilization of memory and, hence, decreased computation time. We also suggest improving the adapted MMA algorithm by adding a frame differencing method as a pre-processing step, then exploiting better multi-scale feature selection to improve average F-score with reference to some of the latest models [35]- [44] on object detection and their post-processing schemes [45]. Furthermore, we are investigating some of the recent deep learning schemes [46]- [57] for detecting and tracking vehicles [27], [34], [39], [58]- [61] in accordance with the complexity analysis [45], [62]- [69] from the deep CNN-based multi-object detection and segmentation schemes [48]- [51], [53]- [56], [59]- [61], [64]- [66], [70]- [72] applied to wide-area aerial surveillance.…”

Section: Conclusion and Further Workmentioning

confidence: 99%

Spatio-Temporal Processing for Automatic Vehicle Detection in Wide-Area Aerial Video

et al. 2020

Self Cite

View full text Add to dashboard Cite

Vehicle detection in aerial videos often requires post-processing to eliminate false detections. This paper presents a spatio-temporal processing scheme to improve automatic vehicle detection performance by replacing the thresholding step of existing detection algorithms with multi-neighborhood hysteresis thresholding for foreground pixel classification. The proposed scheme also performs spatial post-processing, which includes morphological opening and closing to shape and prune the detected objects, and temporal post-processing to further reduce false detections. We evaluate the performance of the proposed spatial processing on two local aerial video datasets and one parking vehicle dataset, and the performance of the proposed spatio-temporal processing scheme on five local aerial video datasets and one public dataset. Experimental evaluation shows that the proposed schemes improve vehicle detection performance for each of the nine algorithms when evaluated on seven datasets. Overall, the use of the proposed spatio-temporal processing scheme improves average F-score to above 0.8, achieves a 90% reduction in false positives and reduces percentage of wrong classifications to below 50% for some of the nine algorithms. INDEX TERMS Vehicle detection, hysteresis thresholding, spatio-temporal processing, wide-area aerial imagery.

show abstract

Performance evaluation of automatic object detection with post-processing schemes under enhanced measures in wide-area aerial imagery

Cited by 6 publications

References 69 publications

A comparative study of intelligent surveillance with big data: methods, strengths, weaknesses, and future directions

A comparative study of intelligent surveillance with big data: methods, strengths, weaknesses, and future directions

An explainable and efficient deep learning framework for video anomaly detection

Spatio-Temporal Processing for Automatic Vehicle Detection in Wide-Area Aerial Video

Contact Info

Product

Resources

About