Sunzil Khandaker

Sunzil Khandaker

Auto-generated profile, based on publicly available information

Sunzil Khandaker

Daffodil International University

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 February 2026 in Smart Agricultural Technology
Verified
Journal Article
Open access
SKSunzil Khandaker
Sunzil Khandaker
MRMd Mizanur Rahman
Md Mizanur Rahman
AAAkm Azad
Akm Azad
MMMohammad Ali Moni
Mohammad Ali Moni
MHMd. Ali Hossain
Md. Ali Hossain

15 July 2025 in Data in Brief
Verified
Journal Article
Open access
SKSunzil Khandaker
Sunzil Khandaker
MRMizanur Rahman
Mizanur Rahman
This extensive dataset of herbal plants can be highly beneficial for the potential development of agricultural research and practical plant identification tasks. This article introduces a dataset named "BDHerbalPlants" with 1792 raw, high-quality images and 14,336 augmented data images of herbal plants. It contains images of eight distinct herbal plants from different regions, including Dhaka and Tangail. The eight plants are Eclipta prostrata, Ocimum tenuiflorum, Centella asiatica, Mentha arvensis, Kalanchoe pinnata, Azadirachta indica, Coriandrum sativum, Datura stramonium. Each image is carefully captured and labeled after verifying by experts. The significance of this dataset is showcased using popular pre-trained models such as Xception, DenseNet201, and RegNetY032 DL models. This herbal plant data has immense potential to be seamlessly integrated into Deep learning (DL) tasks and play a significant role in the healthcare and pharmaceutical industry. It serves as valuable data for future research for agricultural Informatics and classifying herbal plants that can be found in woods but are challenging to identify without field knowledge.

23 March 2026 in Research Square
Verified
Preprint
Open access
PDF file
SKSunzil Khandaker
Sunzil Khandaker
Deep learning image classifiers are increasingly deployed to resource constrained edge devices through INT8 post-training quantization (PTQ), including plant disease imaging. Although quantization reduces memory footprint and can improve inference efficiency, the impact on model explainability is not well comprehended. This study introduces explainability drift, the systematic change in saliency-based explanations between full-precision (FP32) models and quantized (INT8) models, and suggests a multi-metric evaluation suite, including top-k Intersection-over-Union, top-k Dice coefficient, and Spearman rank correlation to measure it. We apply four post-hoc explainability methods (Grad-CAM, Grad-CAM++, Integrated Gradients, and LIME) to Paddy Disease Classification data (10 classes, 10,407 rice leaf images) using three architectures of different complexity (EfficientNetV2-S, ResNet50, and MobileNetV3-Large) before and after INT8 PTQ. Our findings reveal a significant drift depending on both architecture and method. All of the models have an average top-k IoU of less than 0.17, indicating low overlap between the FP32 and INT8 top-saliency regions. Notably, results show that the depthwise-separable architecture of MobileNetV3 leads to a total collapse of Grad-CAM in INT8 with spatially uniform attribution maps (Spearman r = 0.000). In addition, we demonstrate that XAI-preserving quantization-aware training (QAT) with a teacher–student CAM consistency loss can improve Spearman correlation by up to 49.2%, although the results vary depending on the architecture. The work, to the best of our knowledge, provides a quantization-induced explainability drift multi-architecture benchmark and suggests practical recommendations on how to implement reliable, explainable AI in edge devices.