POST PROCESSING OF PREDICTIONS TO IMPROVE THE QUALITY OF RECOGNITION OF WATER SURFACE OBJECTS
DOI:
https://doi.org/10.15588/1607-3274-2024-4-13Keywords:
UAV, detection, recognized objects, water surface, neural network, dataset, model, image distribution, error matrix, training metrics, magnification, image mosaicking, image post-processing, implementation script, mission log, databaseAbstract
Context. The significance of this work stems from the growing need for UAV technologies integrated with artificial intelligence, aimed at detecting and identifying objects on the surface of water bodies. Modern needs in water body monitoring, especially in the context of environmental monitoring, protection and resource management, require accurate and reliable solutions. This work demonstrates methods for improving the performance of neural networks and offers approaches for processing NN predictions, even if they are trained on irrelevant data, which increases the versatility and efficiency of the technology.
Objective. The goal of the work is to solve the problem of false recognition of objects on the surface of water bodies, which is due to a decrease in the accuracy threshold for the neural network. This provides more accurate and reliable detection, reducing the number of false positive predictions and increasing the efficiency of the system in general.
Method. It is proposed to add a stage of post-processing of NN predictions, which inherits concepts of min-max suppression used by YOLO models. This algorithm suppresses the re-detection of the object by the network and relies on the cross-sectional area of the detected rectangles. It uses a threshold value of 0.8 for the two points of the rectangle, which can effectively reduce the number of re-predictions and improve the accuracy.
Results. As a result of the implementation of the proposed algorithm and the script created on its basis, a result was achieved in which groups from several predictions are combined and filtered. The received data is stored in the database as found and detected objects. The proposed post-processing algorithm effectively removes redundant predictions while maintaining forecast accuracy. This ensures the reliability of the system and increases its performance in real conditions.
Conclusions. Detected images of objects on the surface of water bodies are stored in the database in the form of records with unique file name identifiers. After tests with pre-taken images algorithm proved it`s persistence against data duplication scenarios. This increases the efficiency and reliability of the monitoring system, ensuring accurate and timely detection of objects on the surface of water bodies.
References
Schunfelder P., Stebel F., Andreou N., Kunig M. Deep learning-based text detection and recognition on architectural floor plans, Automation in Construction, 2024, № 157. pp. 105–156. DOI: 10. 1016/j.autcon.2023.105156 (date of access: 06.05.2020).
Giakoumoglou N., Pechlivani E. M., Tzovaras D. Generate-paste-blend-detect: Synthetic dataset for object detection in the agriculture domain, Smart Agricultural Technology, 2023, №5, pp. 100–105. DOI:10.1016/j.atech.2023.100258.
Ashourpour M., Azizpour G., Johansen K. Real-time defect and object detection in assembly line: A case for in-line quality inspection, Lecture Notes in Mechanical Engineering, 2024, pp. 99–106. DOI: 10.1007/978-3-031-38241-3_12.
Azevedo P., Santos V. Comparative analysis of multiple yolo-based target detectors and trackers for adas in edge devices, Robotics and Autonomous Systems, 2024, № 171, pp. 87–95. DOI: 10. 1016/j.robot.2023.104558.
Sanjai Siddharthan M., Aravind S., Sountharrajan S. Real-time road hazard classification using object detection with deep learning, Lecture Notes in Networks and Systems, 2024, № 789, LNNS, pp. 479– 492. DOI: 10.1007/978-981-99-6586-1_33.
Wei Z., Zhang Y., Wang X., Zhou J., Dou F., Xia Y. A yolov8-based approach for steel plate surface defect detection, Metalurgija, 2024, № 63, pp. 28–30. DOI: 10.3390/app14125325
Wu F., Zhang Y., Wang L., Hu Q., Fan S., Cai W. A deep learning-based lightweight model for the detection of marine fishes, Journal of Marine Science and Engineering, 2023, №11, pp. 19–34. DOI: 10.3390/jmse11112156.
Zhang G., Tang Y., Tang H., Li W., Wang L. A global lightweight deep learning model for express package detection, Journal of Intelligent and Fuzzy Systems, 2023, № 45. pp. 12013–12025. DOI: 10.3233/JIFS232874.
Wang J., Dai H., Chen T., Liu H., Zhang X., Zhong Q., Lu R. Toward surface defect detection in electronics manufacturing by an accurate and lightweight yolo-style object detector, Scientific Reports, 2023, № 13, pp. 7–26. DOI: 10.1038/s41598-023-33804-w.
Li A., Zhang Z., Sun S., Feng M., Wu C. Multinet-gs: Structured road perception model based on multi-task convolutional neural network, Electronics (Switzerland), 2023, № 12 (19), pp. 39–59. DOI: 10.3390/electronics12193994.
Han L., Ma C., Liu Y., Jia J., Sun J. Sc-yolov8: A security check model for the inspection of prohibited items in x-ray images, Electronics (Switzerland), 2023, № 12 (20), pp. 208–223. DOI: 10.3390/electronics12204208.
Mao J., Wang L., Wang N., Hu Y., Sheng W. A novel method of human identification based on dental impression image, Pattern Recognition, 2023, № 144, pp. 109–126. DOI: 10.1016/j.patcog.2023.109864.
Kara E., Zhang G., Williams J. J., Ferrandez-Quinto G., Rhoden L. J., Kim M., Kutz J. N., Rahman A. Deep learning based object tracking in walking droplet and granular intruder experiments, Journal of RealTime Image Processing, 2023, № 20, pp. 269–311 DOI: 10.1007/s11554-023-01341-4.
Zhou S., Zhong M., Chai X., Zhang N., Zhang Y., Sun Q., Sun T. Framework of rod-like crops sorting based on multi-object oriented detection and analysis, Computers and Electronics in Agriculture, 2024, № 216, pp. 108–145. DOI: 10.1016/j.compag.2023.108516
Shan P., Yang R., Xiao H., Zhang L., Liu Y., Fu Q., Zhao Y. Uavpnet: A balanced and enhanced uav object detection and pose recognition network, Measurement: Journal of the International Measurement Confederation, 2023, № 222, pp. 113–132. DOI: 10.1016/j.measurement.2023.113654.
Talaat F. M., ZainEldin H. An improved fire detection approach based on yolov8 for smart cities, Neural Computing and Applications, 2023, № 35, pp. 20939– 20954. DOI: 10.1007/s00521-023-08809-1.
Liu S., Fan Q., Zhao C., Li S. Rtad: A real-time animal object detection model based on a large selective kernel and channel pruning, Information (Switzerland), 2023, № 14 (10), pp. 535–550 DOI: 10.3390/info14100535.
Smolij V. About features of management preproduction of electronic vehicles, Problems of Modeling and Design Automatization, 2019, № 11, pp. 33–42. DOI: 10.31474/2074-7888-2019-1-33-42.
Su Y., Tan W., Dong Y., Xu W., Huang P., Zhang J., Zhang D. Enhancing concealed object detection in active millimeter wave images using wavelet transform, Signal Processing, 2024, № 216, pp. 109–122. DOI: 10.1016/j.sigpro.2023.109303.
Liu C., Wang K., Li Q., Zhao F., Zhao K., Ma H. Powerful-iou: More straightforward and faster bounding box regression loss with a nonmonotonic focusing mechanism, Neural Networks, 2024, № 170, pp. 276– 284. DOI: 10.1016/j.neunet.2023.11.041.
Xu W., Liu C., Wang G., Zhao Y., Yu J., Muhammad A., Li D. Behavioral response of fish under ammonia nitrogen stress based on machine vision, Engineering Applications of Artificial Intelligence, 2024, № 128, pp. 107–134. DOI: 10.1016/j.engappai.2023.107442.
Dimauro G., Barbaro N., Camporeale M. G., Fiore V., Gelardi M., Scalera M. Deepcilia: Automated, deeplearning based engine for precise ciliary beat frequency estimation, Biomedical Signal Processing and Control, 2024, № 90, pp. 1002–1019. DOI: 10.1016/j.bspc.2023.105808.
Zhao X., Song Y. Improved ship detection with yolov8 enhanced with mobilevit and Gsconv, Electronics (Switzerland), 2023, № 12(22), pp. 46–66. DOI: 10.3390/electronics12224666.
Smolij V. M., Smolij N. V., Sayapin S. P. Search and classification of objects in the zone of reservoirs and coastal zones, CEUR Workshop Proceedings, 2024, No. 3666, pp. 37–51. EID: 2-s2.0-85191443231
Hui J. mAP (mean Average Precision) for Object Detection. Medium. URL: https://jonathanhui.medium.com/map-mean-average-precision-forobject-detection-45c121a31173 (date of access: 15.06.2024).
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 V. M. Smolij, N. V. Smolij, M. V. Mokriiev
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Creative Commons Licensing Notifications in the Copyright Notices
The journal allows the authors to hold the copyright without restrictions and to retain publishing rights without restrictions.
The journal allows readers to read, download, copy, distribute, print, search, or link to the full texts of its articles.
The journal allows to reuse and remixing of its content, in accordance with a Creative Commons license СС BY -SA.
Authors who publish with this journal agree to the following terms:
-
Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License CC BY-SA that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
-
Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
-
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
