Poornima Singh Thakur
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Poornima Singh Thakur has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Aparajita Ojha
PDPM Indian Institute of Information Technology, Design and Manufacturing, Jabalpur, India
9 shared publications - Pritee Khanna
PDPM Indian Institute of Information Technology, Design and Manufacturing, Jabalpur, India
8 shared publications - Shubhangi Chaturvedi
PDPM Indian Institute of Information Technology, Design and Manufacturing, Jabalpur, India
3 shared publications Tanuja Sheorey
Department of Mechanical Engineering (ME), PDPM IIITDM, Jabalpur, Madhya Pradesh, India
3 shared publications- Tanuja Sheorey
Department of Mechanical Engineering, PDPM IIITDM, Jabalpur, India
3 shared publications Shubhangi Chaturvedi
Department of Computer Science and Engineering (CSE), PDPM Indian Institute of Information Technology, Design, and Manufacturing (IIITDM), Jabalpur, Madhya Pradesh, India
2 shared publications
Latest Publications
- 13 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 21 (5) , 3806-3815
Satellite Image-Based Surveillance and Early Wildfire Smoke Detection Using a Multiattention Interlaced Network
The increasing frequency and intensity of wildfires in recent years have not only devastated forest ecosystems, but have also caused a significant economic burden. According to a World Economic Forum report, annual expenditures to combat wildfire hazards is estimated to be more than $ {\$} 50$ billion. This calls for advanced solutions, such as remote sensing surveillance and the use of artificial intelligence for wildfire management. In recent years, several vision-based artificial intelligence techniques have been proposed for fire–smoke image classification that utilise convolutional neural networks. However, challenges persist, particularly in identifying fire–smoke under complex atmospheric conditions. In this article, we introduce a novel multiattention network that interlaces the vision transformer and convolutional neural network to detect fire–smoke in diverse conditions, including clouds, fog, hurricanes, storms, snow, and normal weather. The proposed model not only outperforms eight state-of-the-art fire–smoke image classification methods, but also reduces false alarms by 30% on IIITDMJ $\_$ Smoke dataset and by 6% on UTSC $\_$ SmokeRS dataset. The model also efficiently identifies even tiny occurrence of smoke covering as little as 2% area of an image. The model has also been tested on industrial chimney smoke images and outdoor video fire–smoke scenes. Furthermore, the lightweight architecture of the model with only 0.7 million parameters and 0.2 billion floating point operations per second makes it suitable for deployment on Internet of Things-based forest and industrial surveillance systems.
- 13 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 21 (5) , 3806-3815
Satellite Image-Based Surveillance and Early Wildfire Smoke Detection Using a Multiattention Interlaced Network
The increasing frequency and intensity of wildfires in recent years have not only devastated forest ecosystems, but have also caused a significant economic burden. According to a World Economic Forum report, annual expenditures to combat wildfire hazards is estimated to be more than $ {\$} 50$ billion. This calls for advanced solutions, such as remote sensing surveillance and the use of artificial intelligence for wildfire management. In recent years, several vision-based artificial intelligence techniques have been proposed for fire–smoke image classification that utilise convolutional neural networks. However, challenges persist, particularly in identifying fire–smoke under complex atmospheric conditions. In this article, we introduce a novel multiattention network that interlaces the vision transformer and convolutional neural network to detect fire–smoke in diverse conditions, including clouds, fog, hurricanes, storms, snow, and normal weather. The proposed model not only outperforms eight state-of-the-art fire–smoke image classification methods, but also reduces false alarms by 30% on IIITDMJ $\_$ Smoke dataset and by 6% on UTSC $\_$ SmokeRS dataset. The model also efficiently identifies even tiny occurrence of smoke covering as little as 2% area of an image. The model has also been tested on industrial chimney smoke images and outdoor video fire–smoke scenes. Furthermore, the lightweight architecture of the model with only 0.7 million parameters and 0.2 billion floating point operations per second makes it suitable for deployment on Internet of Things-based forest and industrial surveillance systems.
- Top Cited Papers
- 3 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Biology and Bioinformatics
- Vol. 22 (1) , 310-321
An Ultra Lightweight Interpretable Convolution-Vision Transformer Fusion Model for Plant Disease Identification: ConViTX
Plant diseases have been detrimental for the agriculture industry, as they cause substantial crop loss globally. To overcome this, IoT and AI-based smart agriculture solutions are being deployed for plant disease detection. However, a diverse range of crops and their diseases pose enormous challenges to these methods. Additionally, limited generalizability and the black-box nature of existing deep learning models, together with the scarcity of in-field datasets, are the main bottlenecks in developing efficient and acceptable solutions for large-scale applications. In the present work, a lightweight model ‘ConViTX’ is proposed for plant disease classification that demonstrates improved generalizability and explainability. The compact architecture of ConViTX uses a fusion of convolutional neural networks and vision transformers to simultaneously capture local and global features. Remarkably, ConViTX outperforms nine state-of-the-art deep learning methods on four publicly available datasets and a self-collected in-field maize dataset. Furthermore, the model demonstrates explainable prediction through Gradient Weighted Class Activation Maps and Locally Interpretable Model-Agnostic Evaluations. ConViTX attains 98.8% accuracy on the maize dataset and 61.42% on drone camera-captured raw images. With only 0.7 million parameters and 0.647 billion operations per second, the proposed model has the potential for deployment on resource-constrained precision agriculture setups.
- 3 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Biology and Bioinformatics
- Vol. 22 (1) , 310-321
Top Cited PapersAn Ultra Lightweight Interpretable Convolution-Vision Transformer Fusion Model for Plant Disease Identification: ConViTX
Plant diseases have been detrimental for the agriculture industry, as they cause substantial crop loss globally. To overcome this, IoT and AI-based smart agriculture solutions are being deployed for plant disease detection. However, a diverse range of crops and their diseases pose enormous challenges to these methods. Additionally, limited generalizability and the black-box nature of existing deep learning models, together with the scarcity of in-field datasets, are the main bottlenecks in developing efficient and acceptable solutions for large-scale applications. In the present work, a lightweight model ‘ConViTX’ is proposed for plant disease classification that demonstrates improved generalizability and explainability. The compact architecture of ConViTX uses a fusion of convolutional neural networks and vision transformers to simultaneously capture local and global features. Remarkably, ConViTX outperforms nine state-of-the-art deep learning methods on four publicly available datasets and a self-collected in-field maize dataset. Furthermore, the model demonstrates explainable prediction through Gradient Weighted Class Activation Maps and Locally Interpretable Model-Agnostic Evaluations. ConViTX attains 98.8% accuracy on the maize dataset and 61.42% on drone camera-captured raw images. With only 0.7 million parameters and 0.647 billion operations per second, the proposed model has the potential for deployment on resource-constrained precision agriculture setups.
- Open Access
- 16 September 2024
- Published by Springer Nature in Fire Ecology
- Vol. 20 (1) , 1-17
Ultra-lightweight convolution-transformer network for early fire smoke detection
Background: Forests are invaluable resources, and fire is a natural process that is considered an integral part of the forest ecosystem. Although fire offers several ecological benefits, its frequent occurrence in different parts of the world has raised concerns in the recent past. Covering millions of hectares of forest land, these fire incidents have resulted in the loss of human lives, wild habitats, civil infrastructure, and severe damage to the environment. Around 90% of wildland fires have been caused by humans intentionally or unintentionally. Early detection of fire close to human settlements and wildlife centuries can help mitigate fire hazards. Numerous artificial intelligence-based solutions have been proposed in the past decade that prioritize the detection of fire smoke, as it can be caught through remote sensing and provide an early sign of wildland fire. However, most of these methods are either computationally intensive or suffer from a high false alarm rate. In this paper, a lightweight deep neural network model is proposed for fire smoke detection in images captured by satellites or other remote sensing sources. Results: With only 0.6 million parameters and 0.4 billion floating point operations per second, the hybrid network of convolutional and vision transformer blocks efficiently detects smoke in normal and foggy environmental conditions. It outperforms seven state-of-the-art methods on four datasets, including a self-collected dataset from the “Moderate Resolution Imaging Spectroradiometer” satellite imagery. The model achieves an accuracy of more than 99% on three datasets and 93.90% on the fourth dataset. The t-distributed stochastic neighbor embedding of extracted features by the proposed model demonstrates its superior feature learning capabilities. It is remarkable that even a tiny occurrence of smoke covering just 2% of the satellite image area is efficiently detected by the model. Conclusions: With low memory and computational demands, the proposed model works exceedingly well, making it suitable for deployment in resource constrained devices for forest surveillance and early fire smoke detection.
- 16 September 2024
- Published by Springer Nature in Fire Ecology
- Vol. 20 (1) , 1-17
Open AccessUltra-lightweight convolution-transformer network for early fire smoke detection
Background: Forests are invaluable resources, and fire is a natural process that is considered an integral part of the forest ecosystem. Although fire offers several ecological benefits, its frequent occurrence in different parts of the world has raised concerns in the recent past. Covering millions of hectares of forest land, these fire incidents have resulted in the loss of human lives, wild habitats, civil infrastructure, and severe damage to the environment. Around 90% of wildland fires have been caused by humans intentionally or unintentionally. Early detection of fire close to human settlements and wildlife centuries can help mitigate fire hazards. Numerous artificial intelligence-based solutions have been proposed in the past decade that prioritize the detection of fire smoke, as it can be caught through remote sensing and provide an early sign of wildland fire. However, most of these methods are either computationally intensive or suffer from a high false alarm rate. In this paper, a lightweight deep neural network model is proposed for fire smoke detection in images captured by satellites or other remote sensing sources. Results: With only 0.6 million parameters and 0.4 billion floating point operations per second, the hybrid network of convolutional and vision transformer blocks efficiently detects smoke in normal and foggy environmental conditions. It outperforms seven state-of-the-art methods on four datasets, including a self-collected dataset from the “Moderate Resolution Imaging Spectroradiometer” satellite imagery. The model achieves an accuracy of more than 99% on three datasets and 93.90% on the fourth dataset. The t-distributed stochastic neighbor embedding of extracted features by the proposed model demonstrates its superior feature learning capabilities. It is remarkable that even a tiny occurrence of smoke covering just 2% of the satellite image area is efficiently detected by the model. Conclusions: With low memory and computational demands, the proposed model works exceedingly well, making it suitable for deployment in resource constrained devices for forest surveillance and early fire smoke detection.
- 16 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on AgriFood Electronics
- Vol. 2 (2) , 576-586
Real-Time Plant Disease Identification: Fusion of Vision Transformer and Conditional Convolutional Network With C3GAN-Based Data Augmentation
Climate change, adverse weather conditions, and illegitimate farming practices have caused severe damage to the agricultural ecosystem, resulting in significant crop loss in the last decade. One of the major challenges is the breakout of plant diseases that harm the crop in the field. To address this issue, several artificial intelligence and Internet of Things-based systems have been developed for crop monitoring and containment of plant diseases at early stages. In this article, a real-time plant disease identification system is designed using drone-based surveillance and farmer's input. A lightweight plant disease classification model is deployed in the proposed system using a fusion of a vision transformer and a convolutional neural network. The proposed model deploys conditional attention with a statistical squeeze-and-excitation module to efficiently learn the plant disease patterns from images captured under normal and challenging weather conditions. With only 0.95 million trainable parameters, the performance of the proposed plant disease classification model surpasses that of seven state-of-the-art techniques on five public datasets and an in-house developed maize dataset from drone camera-captured images under varying environmental conditions. To provide a better learning experience of real-world data to the model, a generative adversarial network, C3GAN, inspired by cycleGAN, is proposed for data augmentation of the collected maize dataset. The system keeps updating the model parameters based on the feedback of agriculture experts and farmers when new diseases break out or the model's performance deteriorates on unseen data during the surveillance over a period of time.
- 16 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on AgriFood Electronics
- Vol. 2 (2) , 576-586
Real-Time Plant Disease Identification: Fusion of Vision Transformer and Conditional Convolutional Network With C3GAN-Based Data Augmentation
Climate change, adverse weather conditions, and illegitimate farming practices have caused severe damage to the agricultural ecosystem, resulting in significant crop loss in the last decade. One of the major challenges is the breakout of plant diseases that harm the crop in the field. To address this issue, several artificial intelligence and Internet of Things-based systems have been developed for crop monitoring and containment of plant diseases at early stages. In this article, a real-time plant disease identification system is designed using drone-based surveillance and farmer's input. A lightweight plant disease classification model is deployed in the proposed system using a fusion of a vision transformer and a convolutional neural network. The proposed model deploys conditional attention with a statistical squeeze-and-excitation module to efficiently learn the plant disease patterns from images captured under normal and challenging weather conditions. With only 0.95 million trainable parameters, the performance of the proposed plant disease classification model surpasses that of seven state-of-the-art techniques on five public datasets and an in-house developed maize dataset from drone camera-captured images under varying environmental conditions. To provide a better learning experience of real-world data to the model, a generative adversarial network, C3GAN, inspired by cycleGAN, is proposed for data augmentation of the collected maize dataset. The system keeps updating the model parameters based on the feedback of agriculture experts and farmers when new diseases break out or the model's performance deteriorates on unseen data during the surveillance over a period of time.
- 6 August 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (31) , 19499-19514
Smoke detection in foggy surveillance environment using parallel vision transformer network
- 6 August 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (31) , 19499-19514
Smoke detection in foggy surveillance environment using parallel vision transformer network
- 15 April 2023
- Published by Springer Nature
- p. 308-319
Attention Based Twin Convolutional Neural Network with Inception Blocks for Plant Disease Detection Using Wavelet Transform
- 15 April 2023
- Published by Springer Nature
- p. 308-319
Attention Based Twin Convolutional Neural Network with Inception Blocks for Plant Disease Detection Using Wavelet Transform
- 24 July 2022
- Published by Springer Nature
- p. 338-349
Analysis of Loss Functions for Image Reconstruction Using Convolutional Autoencoder
- 24 July 2022
- Published by Springer Nature
- p. 338-349
Analysis of Loss Functions for Image Reconstruction Using Convolutional Autoencoder
- 24 July 2022
- Published by Springer Nature
- p. 501-511
Vision Transformer for Plant Disease Detection: PlantViT
- 24 July 2022
- Published by Springer Nature
- p. 501-511
Vision Transformer for Plant Disease Detection: PlantViT
- Top Cited Papers
- 11 July 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 208
Trends in vision-based machine learning techniques for plant disease identification: A systematic review
- 11 July 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 208
Top Cited PapersTrends in vision-based machine learning techniques for plant disease identification: A systematic review
- Top Cited Papers
- 7 June 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (1) , 497-520
VGG-ICNN: A Lightweight CNN model for crop disease identification
- 7 June 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (1) , 497-520
Top Cited PapersVGG-ICNN: A Lightweight CNN model for crop disease identification