Knowledge guided deep learning for general-purpose computer vision applications

Djenouri, Youcef; Belbachir, Ahmed Nabil; Jhaveri, Rutvij H.; Djenouri, Djamel

Abstract

This research targets general-purpose smart computer vision that eliminates reliance on domain-specific knowledge to reach adaptable generic models for flexible applications. It proposes a novel approach in which several deep learning models are trained for each image. Statistical information of each trained image is then calculated and stored with the loss values of each model used in the training phase. The stored information is finally used to select the appropriate model for each new image data in the testing phase. To efficiently select the appropriate model, a kNN (k Nearest Neighbors) strategy is used to select the best model in the testing phase. The developed framework called KGDL (Knowledge Guided Deep Learning) was evaluated and tested using two computer vision benchmarks, 1) ImageNet for image classification, and 2) COCO for object detection. The results reveal the effectiveness of KGDL in terms of accuracy and competitiveness of inference runtime. In particular, it achieved 94 % of classification rate in ImageNet, and 92% of intersection over union in COCO dataset.

Citation

Djenouri, Y., Belbachir, A. N., Jhaveri, R. H., & Djenouri, D. (2023). Knowledge guided deep learning for general-purpose computer vision applications. In Computer Analysis of Images and Patterns (185-194). https://doi.org/10.1007/978-3-031-44237-7_18

• Conference Name: International Conference on Computer Analysis of Images and Patterns
• Acceptance Date: Aug 1, 2023
• Online Publication Date: Sep 20, 2023
• Publication Date: Sep 20, 2023
• Deposit Date: Oct 5, 2023
• Publicly Available Date: Sep 21, 2024
• Publisher: Springer Verlag
• Volume: 14184 LNCS
• Pages: 185-194
• Book Title: Computer Analysis of Images and Patterns
• ISBN: 9783031442360
• DOI: https://doi.org/10.1007/978-3-031-44237-7_18
• Public URL: https://uwe-repository.worktribe.com/output/11152187
• Additional Information: First Online: 20 September 2023; Conference Acronym: CAIP; Conference Name: International Conference on Computer Analysis of Images and Patterns; Conference City: Limassol; Conference Country: Cyprus; Conference Year: 2023; Conference Start Date: 25 September 2023; Conference End Date: 28 September 2023; Conference Number: 20; Conference ID: caip2023; Conference URL: https://cyprusconferences.org/caip2023/; Type: Single-blind; Conference Management System: https://www.easyacademia.org; Number of Submissions Sent for Review: 67; Number of Full Papers Accepted: 54; Number of Short Papers Accepted: 0; Acceptance Rate of Full Papers: 81% - The value is computed by the equation "Number of Full Papers Accepted / Number of Submissions Sent for Review * 100" and then rounded to a whole number.; Average Number of Reviews per Paper: 2.06; Average Number of Papers per Reviewer: 2.09; External Reviewers Involved: No