Thesis Defence: Akram Kazemisisi (Master of Science in Computer Science)

Date
to
Location
Senate Chambers and/or Zoom
Campus
Prince George
Online

You are encouraged to attend the defence. The details of the defence and attendance information is included below:  

Date: June 13, 2025
Time: 2:00 PM to 4:00 PM (PT)

Defence mode: Hybrid 
In-Person Attendance: Senate Chambers, UNBC Prince George Campus  
Virtual Attendance: via Zoom 

LINK TO JOIN: Please contact the Office of Graduate Administration for information regarding remote attendance for online defences.   

 To ensure the defence proceeds with no interruptions, please mute your audio and video on entry and do not inadvertently share your screen. The meeting will be locked to entry 5 minutes after it begins: please ensure you are on time.  

Thesis entitled: SO-DRCNN WITH TERNION PARADIGM EXTRACTION ROUTINE FOR AN EFFECTIVE IMAGE RETRIEVAL SYSTEM

Abstract: This thesis addresses the challenges of semantic image retrieval and labeled data scarcity in Content-Based Image Retrieval (CBIR) by introducing SO-DRCNN, a novel Self-Optimizing DeepRec Convolutional Neural Network framework. SO-DRCNN leverages a hybrid approach, combining the strengths of handcrafted features (Ternion Paradigm: HOG, ICH, SERC) and deep learning. A pre-trained ResNet-50 backbone, enhanced with Recurrent Patching (Bi-LSTM), Spatial Pyramid Pooling (SPP/ASPP), and Attention mechanisms, extracts high-level semantic features. A key innovation is the Siamese-Driven Feature Fusion, where a Siamese network, trained with a contrastive loss, learns to adaptively combine handcrafted and deep features, optimizing the fused representation for similarity. This self-supervised training strategy (Auto-Embedder) eliminates the need for manual image labels. Experiments on benchmark datasets demonstrate that SO-DRCNN achieves state-of-the-art retrieval accuracy, outperforming traditional methods and demonstrating the effectiveness of the learned fusion strategy. The system is also integrated with Elasticsearch for scalable retrieval. This work contributes a robust, efficient, and interpretable solution for semantic CBIR.

Defence Committee:  
Chair: Dr. Thomas Tannert, University of Northern British Columbia  
Supervisor: Dr. Liang Chen, University of Northern British Columbia  
Committee Member: Dr. Fan Jiang, University of Northern British Columbia  
Committee Member: Dr. Jianbing Li, University of Northern British Columbia  
External Examiner: Dr. Mohammad El Smaily, University of Northern British Columbia  

Contact Information

Graduate Administration in the Office of the Registrar, University of Northern British Columbia