Real-Time Object Detection via Cloud-Enabled Deep Learning: A Systematic Review | Chapter 6 | Mathematics and Computer Science: Research Updates Vol. 6

 

Automated vehicles, advanced surveillance systems, AR, and robots are just a few of the many new uses for real-time object recognition.  While deep learning models are becoming increasingly complex and accurate, they might be challenging to execute on edge devices with limited resources due to the computational demands. By offloading computationally intensive processes to scalable cloud infrastructure, cloud-enabled deep learning enables real-time processing without sacrificing detection accuracy, offering an effective alternative.  This study takes a close look at the current setup of cloud-based object recognition methods that work in real time. When considering latency, bandwidth, privacy, and processing costs, the pros and cons of several architectural paradigms are evaluated, including hybrid methodology, distributed inference, and edge-cloud cooperation. Additionally, the developments of lightweight convolutional neural networks (CNNs), single-shot detectors, and model compression techniques are examined, all of which are aimed at real-time performance in cloud environments.  Improving fault tolerance, optimizing data transmission, safeguarding data security and privacy, and developing more adaptive and efficient cloud resource management strategies for dynamic real-time object detection contexts are all areas that could be further explored in this review.

 

 

Author(s) Details

Abdul Razzak Khan Qureshi
Department of Computer Science, Medicaps University, Indore, Madhya Pradesh, India.

 

Ruby Bhatt
Department of Computer Science, Medicaps University, Indore, Madhya Pradesh, India.

 

Govinda Patil
Department of Computer Science, Medicaps University, Indore, Madhya Pradesh, India.

 

Please see the book here :- https://doi.org/10.9734/bpi/mcsru/v6/5867

 

CNN-Based Prediction of Oxygen Supply Needs in under-Actuated HVAC Zones | Chapter 6 | Engineering Research: Perspectives on Recent Advances Vol. 11

 

Aims: This study aims to develop a predictive model for oxygen consumption in under-actuated HVAC zones by integrating occupant-centric parameters such as metabolic rate and activity behaviour with environmental and mechanical variables.

 

Study Design: This is an experimental and computational modelling study utilising a dataset of Variable Air Volume (VAV) damper measurements, fan energy estimation, and occupant metabolic rate calculations.

 

Place and Duration of Study: The dataset was collected and processed at Universitas Trilogi, Jakarta, Indonesia, during 2024–2025.

 

Methodology: A curated dataset comprising 11,696 multivariate records was used, including occupant activity counts, environmental parameters (temperature, humidity, air pressure), HVAC operational metrics (airflow, fan speed, brake horsepower, damper percentage), and physiological measures (metabolic rate, daily and minute-based kilocalorie expenditure). Oxygen consumption (L/min) was computed using metabolic energy equivalence. A Convolutional Neural Network (CNN) model was developed and benchmarked against a Multi-Layer Perceptron Artificial Neural Network (MLP-ANN) and a standard Multi-Layer Perceptron (MLP) Regressor. Model performance was assessed using RMSE, MAE, and R² metrics.

 

Results: Descriptive statistics indicated significant variability in metabolic rate and oxygen demand, ranging from near zero during sedentary activities to peaks above 0.9 L/min during high-intensity activities. The CNN model achieved the best performance (RMSE = 0.0074, MAE = 0.0038, R² = 0.9923), outperforming MLP-ANN (RMSE = 0.0314, MAE = 0.0151, R² = 0.8733) and other benchmarks. Learning curves demonstrated stable convergence without overfitting, while parity plots confirmed strong alignment between predicted and observed oxygen consumption.

 

Conclusion: The findings confirm that occupant behaviour, particularly metabolic rate, plays a critical role in predicting oxygen demand in under-actuated zones. CNN demonstrated superior predictive accuracy and generalisation by effectively capturing nonlinear and spatio-temporal dependencies. These results suggest that HVAC systems can be optimised through adaptive oxygen control strategies, enhancing both energy efficiency and indoor air quality.

 

Author(s) Details

Yaddarabullah
Department of Informatics, Universitas Trilogi, Indonesia.

 

Erneza Dewi Krishnasari
Department of Visual Communication Design, Universitas Trilogi, Indonesia.

 

Idea Alvira
Department of Tourism, Institute Seni Indonesia Padangpanjang, Indonesia.

 

Please see the book here :- https://doi.org/10.9734/bpi/erpra/v11/6530

A Comprehensive Review of Advanced AI/ML Applications in Solar Radiation Prediction | Chapter 9 | Scientific Research, New Technologies and Applications Vol. 7

Recent advancements in AI/ML have focused on improving the robustness of solar radiation predictions through hybrid models, combining machine learning algorithms with traditional physical models, or blending multiple ML techniques. Solar energy is one of the most abundant renewable energy sources available globally, and accurate solar radiation prediction is crucial for efficient energy management, grid optimization, and the development of solar technologies. The advent of machine learning and artificial intelligence has revolutionized the field of solar radiation forecasting, offering more accurate, scalable, and adaptive models. In India, using AI/ML for solar radiation prediction is challenged by inconsistent data quality due to limited, fragmented measurement stations across diverse climates. Large-scale, high-quality datasets are often unavailable, especially in remote or rural areas. The high computational resource requirements further strain infrastructure, particularly in less developed regions. Additionally, interpreting AI/ML models in this context can be difficult, limiting trust in predictions and wider adoption across the country. This review provides a comprehensive overview of the state-of-the-art ML and AI approaches used for solar radiation prediction, including traditional models such as regression and time-series analysis, as well as advanced techniques like deep learning and ensemble models. Additionally, the article discusses the challenges, limitations, and future directions in this field. The future of ML and AI in solar radiation prediction looks promising, with potential advancements in quantum computing and the integration of edge computing for real-time forecasting.

 

Author (s) Details

 

R. Meenal
V.S.B Engineering College, Karur - 639111, TN, India.

 

R. Devprakash
Lincoln University College, Petaling Jaya, 47301, Malaysia.

 

E. Rajasekaran
V.S.B Engineering College, Karur - 639111, TN, India.

 

Please see the book here:- https://doi.org/10.9734/bpi/srnta/v7/2935

Analysing the Classification of Rice Leaf Diseases on the Basis of Deep Convolutional Neural Network Architectures | Chapter 9 | Research Highlights in Science and Technology Vol. 3

 In the field of farming, timely investigation and acknowledgment of plant leaf diseases assures extreme crop quality and yield. Due to a lack of knowledge about ultimate cutting-edge refined approaches in the field of leaf disease discovery, one of the largest impediments for rice growers is the identification of leaf diseases. Due to the repetitiveness of rice leaf afflictions, a large portion of rice development is disrupted. Early discovery of rice leaf diseases is immediately done manually by farmers, that is extremely late and labor-intensive. However, the requirement of mechanical disease discovery in rice leaves aids producers in more effectively preserving their land harvests. In this review, the major focus act performance analysis of discovery of rice leaf ailments based on the architectures employed. Convolutional affecting animate nerve organs networks are the best method for classifying edible grain leaf diseases, and advances in calculating vision and deep learning placate predictions and come to a close the greatest method for achievement so. Numerous CNN architectures have been resolved for finding best classification depiction based on training from the very beginning, fine tuning or through transfer knowledge. Here, right selection of Deep CNN architectures for classification purposes supports high performance rates established the type of learning employed.

Author(s) Details:

Taruna Sharma,
Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India.

Puninder Kaur,
Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India.

Jasmeen Chahal,
Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India.

Himanshu Sharma,
Department of Electronics and Communication Engineering, JBIET, Hyderabad, India.

Please see the link here: https://stm.bookpi.org/RHST-V3/article/view/10760

Development of an Automated System for Identification of Skeletal Maturity using Convolutional Neural Networks Approach | Chapter 6 | Technological Innovation in Engineering Research Vol. 7

 In this paper, an advanced automation skeletal recognition system that forecasts bone ageing from a radiograph of the left hand, wrist, and fingers is presented. A faster R-CNN uses the left-hand radiograph as input and outputs the DRU region that was discovered. Since the DRU area occupies the majority of the left-hand region, it aids in determining the bone maturity of newborns and young children and forecasts the puberty's accelerating and retarding stages. The experimental section details the setup of the 1101 radiographs of the left hand and wrist as well as the performance of the model using different optimization techniques and training sample sizes.  The recommended approach finally achieves 92 percent (radius) and 90 percent (ulna) classification accuracy after testing parameter adjustment.

Author(s) Details:

B. Sowmya Reddy,
Sreenidhi Institute of Science and Technology, Affiliated to Jawaharlal Nehru Technical University Hyderabad, Telangana, India.

Devavarapu Sreenivasarao,
Sreenidhi Institute of Science and Technology, Affiliated to Jawaharlal Nehru Technical University Hyderabad, Telangana, India.

Shaik Khasim Saheb,
Sreenidhi Institute of Science and Technology, Affiliated to Jawaharlal Nehru Technical University Hyderabad, Telangana, India.

Please see the link here: https://stm.bookpi.org/TIER-V7/article/view/7921

A Novel Approach for Predicting Disease in Plant Using Hybrid Deep Convolutional Neural Network | Chapter 5 | Current Approaches in Science and Technology Research Vol. 8

 Agriculture is India's most important economic industry. The detection of plant diseases is a key concern in the agriculture industry. An accurate and speedier identification of plant diseases, resulting in significant reductions in economic losses. Manually monitoring plant leaf disease is a highly important task that takes a long time. As a result, an automated solution is required for plant disease detection. Deep learning is quickly becoming the industry standard for picture classification. Researchers have been able to improve the accuracy of object detection thanks to recent advances in Deep Neural Networks. Researchers have already built a few architectures for effective plant disease classification, including the Faster Region-based Convolutional Neural Network (Faster R-CNN), Region-based Fully Convolutional Network (R-FCN), and Single Shot Multibox Detector (SSD). The well-known Deep Convolutional Neural Network architectures for generic Image categorization are Alex Net, Google Net, and VGG-16. When many diseases afflict the same leaf, however, this architecture does not perform well.

To address this, this study proposes a Hybrid Deep Convolutional Neural Network architecture with segmentation, which consists of five convolutional layers, five pooling layers, and two fully connected layers. Deep convolutional neural networks are extensively employed to assess visual imagery and are frequently utilised in image categorization behind the scenes. The input photos are fed to the CNN in the proposed Faster-RCNN system, increasing the accuracy of images to forecast illnesses from plant leaves.

Author (s) Details

R. Jayavadivel
Department of Computer Science and Engineering, Lovely Professional University, Jalandhar-Delhi, G. T. Road, Phagwara, Punjab, India.

V. Chandrasekar
Department of Computer Science and Engineering, Malla Reddy College of Engineering and Technology, Secunderabad, Telangana State, India.

V. Shanmugavalli
Department of Computer Science and Engineering, Vivekanandha College of Engineering for Women, India.

View Book :- https://stm.bookpi.org/CASTR-V8/article/view/1946