Joyanta Jyoti Mondal

Joyanta Jyoti Mondal

Auto-generated profile, based on publicly available information

Joyanta Jyoti Mondal

Department of Computer and Information Sciences University of Delaware United States

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
29 June 2022
Verified
Conference Paper
Open access
JMJoyanta Jyoti Mondal
Joyanta Jyoti Mondal
FIFarhadul Islam
Farhadul Islam
SZSarah Zabeen
Sarah Zabeen
AIA. B. M. Alim Al Islam
A. B. M. Alim Al Islam
JNJannatun Noor
Jannatun Noor
Agriculture is the fundamental source of revenue and Gross Domestic Product (GDP) in many countries where economically developing countries; especially the Global South are no exception. Various types of plant-based diseases are strongly intertwined with the everyday lives of those who are connected with agriculture. Among the diseases, most of them can be diagnosed by leaves. However, due to the variety of illnesses, identifying and classifying any plant leaf disease is difficult and time-consuming. Besides, late identifications of diseases cause losses for the farmers on a large scale, which in turn affects their financial state. Therefore, to overcome this problem, we present a lightweight approach (called PLeaD-Net) to accurately recognize and categorize plant leaf diseases in this paper. Here, leveraging a limited-resource deep convolutional network (Deep CNN) model, we extract information from sick sections of a leaf to accurately identify locations of disease. In comparison to existing deep learning methods and other prior research, our proposed approach achieves a much higher performance using fewer parameters as per our experimental results. In our study and experimentation, we develop and implement an architecture based on Deep CNN. We test our architecture on a publicly available dataset that contains different types of plant leaves images and backgrounds.

23 October 2024 in Engineering Applications of Artificial Intelligence
Verified
Journal Article
Closed access
FIFarhadul Islam
Farhadul Islam
MMMeem Arafat Manab
Meem Arafat Manab
JMJoyanta Jyoti Mondal
Joyanta Jyoti Mondal
SZSarah Zabeen
Sarah Zabeen
FRFardin Bin Rahman
Fardin Bin Rahman
JNJannatun Noor
Jannatun Noor

28 March 2022 in IEEE Transactions on Cloud Computing
Verified
Journal Article
Closed access
JNJannatun Noor
Jannatun Noor
NSNazrul Huda Shanto
Nazrul Huda Shanto
JMJoyanta Jyoti Mondal
Joyanta Jyoti Mondal
GHGolam Hossain
Golam Hossain
SCSriram Chellappan
Sriram Chellappan
AIA. B. M. Alim Al Islam
A. B. M. Alim Al Islam
Since massive numbers of images are now being communicated from, and stored in different cloud systems, faster retrieval has become extremely important. This is more relevant, especially after COVID-19 in bandwidth-constrained environments. However, to the best of our knowledge, a coherent solution to overcome this problem is yet to be investigated in the literature. In this paper, by customizing the Progressive JPEG method, we propose a new Scan Script to ensure Faster Image Retrieval. Furthermore, we also propose a new lossy PJPEG architecture to reduce the file size as a solution to overcome our Scan Script's drawback. In order to achieve an orchestration between them, we improve the scanning of Progressive JPEG's picture payloads to ensure Faster Image Retrieval using the change in bit pixels of distinct Luma and Chroma components (Y, Cb, and Cr). The orchestration improves user experience even in bandwidth-constrained cases. We evaluate our proposed orchestration in a real-world setting across two continents encompassing a private cloud. Compared to existing alternatives, our proposed orchestration can improve user waiting time by up to 54% and decrease image size by up to 27%. Our proposed work is tested in cutting-edge cloud apps, ensuring up to 69% quicker loading time.

28 March 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
SLS. M. Fazle Rabby Labib
S. M. Fazle Rabby Labib
JMJoyanta Jyoti Mondal
Joyanta Jyoti Mondal
MMMeem Arafat Manab
Meem Arafat Manab
XXXi Xiao
Xi Xiao
SNSarfaraz Newaz
Sarfaraz Newaz
The susceptibility of deep neural networks (DNNs) to adversarial attacks undermines their reliability across numerous applications, underscoring the necessity for an in-depth exploration of these vulnerabilities and the formulation of robust defense strategies. The DeepFool algorithm by Moosavi-Dezfooli et al. (2016) represents a pivotal step in identifying minimal perturbations required to induce misclassification of input images. Nonetheless, its generic methodology falls short in scenarios necessitating targeted interventions. Additionally, previous research studies have predominantly concentrated on the success rate of attacks without adequately addressing the consequential distortion of images, the maintenance of image quality, or the confidence threshold required for misclassification. To bridge these gaps, we introduce the Enhanced Targeted DeepFool (ET DeepFool) algorithm, an evolution of DeepFool that not only facilitates the specification of desired misclassification targets but also incorporates a configurable minimum confidence score. Our empirical investigations demonstrate the superiority of this refined approach in maintaining the integrity of images and minimizing perturbations across a variety of DNN architectures. Unlike previous iterations, such as the Targeted DeepFool by Gajjar et al. (2022), our method grants unparalleled control over the perturbation process, enabling precise manipulation of model responses. Preliminary outcomes reveal that certain models, including AlexNet and the advanced Vision Transformer, display commendable robustness to such manipulations. This discovery of varying levels of model robustness, as unveiled through our confidence level adjustments, could have far-reaching implications for the field of image recognition. Our code is available at https://github.com/FazleLabib/et_deepfool

14 May 2024 in IEEE Access
Verified
Journal Article
Open access
JNJannatun Noor
Jannatun Noor
SSSadiqul Islam Sakif
Sadiqul Islam Sakif
JMJoyanta Jyoti Mondal
Joyanta Jyoti Mondal
MUMir Rownak Ali Uday
Mir Rownak Ali Uday
RRRizwanul Haque Ratul
Rizwanul Haque Ratul
AIA. B. M. Alim Al Islam
A. B. M. Alim Al Islam
Cloud-based Object Storage Systems (OSS) are known for their scalability, durability, availability, and concurrency. However, there is a noticable vaccum in open-source OSS for a straightforward way for users and administrators to conduct data searches within object storage without fully utilizing the cloud infrastructure. In our research, we present Sherlock, a novel Content-Based Searching (CoBS) framework. Sherlock enhances search capabilities by using extra information from images and documents, incorporating this information into an Elasticsearch-powered database to enable content-driven searches. The framework operates through a two-stage process. First, it classifies the incoming data by type, directing images to an object detection model and processing documents for keyword extraction. Then, Elasticsearch catalogs the extracted data, facilitating searches based on content. The effectiveness of our searches is largely dependent on the precision of these models, which we improve by training them on large-scale datasets: the Microsoft COCO Dataset for multimedia content and the SemEval2017 Dataset for text documents. We further test our system’s performance by integrating it with the open-source OSS, OpenStack Swift, and conducting real-world experiments with image uploads to evaluate how our model performs within Swift’s object storage environments.

16 August 2023
Verified
Conference Paper
Open access
AAAditto Baidya Alok
Aditto Baidya Alok
HSHasibul Sakib
Hasibul Sakib
SUShamsil Arafin Ullah
Shamsil Arafin Ullah
FHFardin Huq
Fardin Huq
RGRiya Ghosh
Riya Ghosh
JNJannatun Noor
Jannatun Noor
With rapid urbanization and increasing traffic congestion in major cities, alternative modes of transportation have gained significant attention. The ride-sharing app revolution also has sparked a significant transformation in the transportation industry, along with “Khep”, an unusual ride-sharing approach where individuals negotiate fares directly with drivers personally, which has become a popular means of transportation. In this research, we investigate various factors that influence the preference for such unconventional contractual rides over ride-sharing apps, such as cultural norms, trust issues, affordability, and accessibility. Moreover, we explore the role of technology literacy, marketing strategies, and regulatory frameworks in shaping the adoption landscape. After conducting a survey and interviews, we conduct a thorough analysis to determine the expected findings and uncover meaningful insights regarding the utilization, preferences, and challenges within the ride-sharing industry in Bangladesh. The findings of this study reveal that cultural factors, such as the preference for bargaining and personal connections, strongly influence the popularity of contractual rides. Additionally, concerns related to safety, data privacy, and trust act as barriers to ride-sharing app adoption. Being associated with UN SDG Goals 11 and 9, the implications of this research extend beyond Bangladesh and can provide insights for policymakers, transportation companies, and technology developers seeking to understand the factors shaping the adoption of ride-sharing apps in similar contexts.