In order for a country’s economy to grow, agricultural development is essential. Plant diseases, however, severely hamper crop growth rate and quality. In the absence of domain experts and with low contrast information, accurate identification of these diseases is very challenging and time-consuming. This leads to an agricultural management system in need of a method for automatically detecting disease at an early stage. As a consequence of dimensionality reduction, CNN-based models use pooling layers, which results in the loss of vital information, including the precise location of the most prominent features. In response to these challenges, we propose a fine-tuned technique, GreenViT, for detecting plant infections and diseases based on Vision Transformers (ViTs). Similar to word embedding, we divide the input image into smaller blocks or patches and feed these to the ViT sequentially. Our approach leverages the strengths of ViTs in order to overcome the problems associated with CNN-based models. Experiments on widely used benchmark datasets were conducted to evaluate the proposed GreenViT performance. Based on the obtained experimental outcomes, the proposed technique outperforms state-of-the-art (SOTA) CNN models for detecting plant diseases.
29 November 2023 in IEEE Internet of Things Journal
Advancements in deep learning and the Internet of Things (IoT) enable early fire detection through vision-based systems, reducing ecological, social, and economic damage. These systems necessitate lightweight, cost-effective convolutional neural networks (CNNs) for real-time operation. Effective deployment on AI-assisted edge devices is crucial for optimal performance. To mitigate this problem, we present the optimized fire attention network (OFAN) for effective and efficient fire detection. In the re-engineered attention block, we swapped the convolution layers by dilated variants and integrated additional dense layers to capture global context and refine more weight optimization. We calibrate the OFAN for real-time processing using a lightweight and efficient feature extractor backbone model. Additionally, a challenging fire dataset is a critical contribution that contains extremely diverse, blazing, and non-fire, captured in lighting and foggy environments. It advances traditional fire detection samples by considering low-light and foggy conditions. A comprehensive experiment is conducted over three widely used fire detection datasets, and our proposed OFAN outperforms state-of-the-art. The proposed OFAN achieved 96.23%, 96.54% and 94.63% accuracies over BoWFire, FD and the newly proposed DiverseFire dataset, respectively. Our research sets a standard for fire detection over edge devices, offering improved accuracy and better frames per second (FPS) performance.
Object detection supported by unmanned aerial vehicles (UAVs) has generated significant interest in recent years including applications, such as surveillance, search for missing persons, traffic, and disaster management. Location awareness is a challenging task, particularly, the deployment of UAVs in a global positioning system (GPS) restricted environment or GPS sensor failure. To mitigate this problem, we propose LocateUAV, a novel location awareness framework, to detect UAV’s location by processing the data from the visual sensor in real time using a lightweight convolutional neural network (CNN). Assuming that the drone is in an IoT environment, first, the object detection technique is applied to detect the object of interest (OOI), namely, signboard. Subsequently, optical character recognition (OCR) is applied to extract useful contextual information. In the final step, the extracted information is forwarded to the map application programming interface (API) to locate the UAV. We also present a newly created data set for LocateUAV, which comprises challenging scenarios for context analysis. Moreover, we also compress an existing lightweight model up to 45 MB for efficient processing over UAV, which is 19.5% when compared with the size of the original model. Finally, an in-depth comparison of various trained and efficient object detection and OCR techniques is presented to facilitate future research on the development of flex drone that can extract information from the surroundings of a location in a GPS-restricted environment.
1 April 2024 in Digital Communications and Networks
Sumoylation is a post-translation modification (PTM) mechanism that involves many critical biological processes, such as gene expression, localizing and stabilizing proteins, and replicating the genome. Moreover, sumoylation sites are associated with different diseases, including Parkinson’s and Alzheimer’s. Due to its vital role in the biological process, identifying sumoylation sites in proteins is significant for monitoring protein functions and discovering multiple diseases. Therefore, in the literature, several computational models utilizing conventional ML methods have been introduced to classify sumoylation sites. However, these models cannot accurately classify the sumoylation sites due to intrinsic limitations associated with the conventional learning methods. This paper proposes a robust computational model (called Deep-Sumo) for predicting sumoylation sites based on a deep-learning algorithm with efficient feature representation methods. The proposed model employs a half-sphere exposure method to represent protein sequences in a feature vector. Principal Component Analysis is applied to extract discriminative features by eliminating noisy and redundant features. The discriminant features are given to a multilayer Deep Neural Network (DNN) model to predict sumoylation sites accurately. The performance of the proposed model is extensively evaluated using a 10-fold cross-validation test by considering various statistical-based performance measurement metrics. Initially, the proposed DNN is compared with the traditional learning algorithm, and subsequently, the performance of the Deep-Sumo is compared with the existing models. The validation results show that the proposed model reports an average accuracy of 96.47%, with improvement compared with the existing models. It is anticipated that the proposed model can be used as an effective tool for drug discovery and the diagnosis of multiple diseases.
The recognition of fire at its early stages and stopping it from causing socioeconomic and environmental disasters remains a demanding task. Despite the availability of convincing networks, there is a need to develop a lightweight network for resource-constraint devices rather than real-time fire detection in smart city contexts. To overcome this shortcoming, we presented a novel efficient lightweight network called FlameNet for fire detection in a smart city environment. Our proposed network works via two main steps: first, it detects the fire using the FlameNet; then, an alert is initiated and directed to the fire, medical, and rescue departments. Furthermore, we incorporate the MSA module to efficiently prioritize and enhance relevant fire-related prominent features for effective fire detection. The newly developed Ignited-Flames dataset is utilized to undertake a thorough analysis of several convolutional neural network (CNN) models. Additionally, the proposed FlameNet achieves 99.40% accuracy for fire detection. The empirical findings and analysis of multiple factors such as model accuracy, size, and processing time prove that the suggested model is suitable for fire detection.