G. Vennila

G. Vennila

Auto-generated profile, based on publicly available information

G. Vennila

Mohan Babu University, India

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

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

Representative Publications

View All Publications
30 June 2026
Verified
Book Chapter
Closed access
GVG. Vennila
G. Vennila
BSB. Prabhu Shankar
B. Prabhu Shankar
SAS. Anitha
S. Anitha
SVS. Vijayalakshmi
S. Vijayalakshmi
Music genre classification is essential in music retrieval and recommendation systems. This work uses an Inception-based Recurrent Neural Network (Inception-RNN) for automatic genre identification. Raw audio is converted into mel-spectrograms, which act as model inputs. The modified Inception modules extract multi-scale frequency patterns, learning rich and discriminative local features. These are passed to GRU layers that capture temporal dependencies and the evolving structure of sound over time. By combining multi-scale spectral analysis with sequential modeling, the framework effectively handles both short-term audio cues and long-term musical patterns. The method shows strong performance for genres with notable temporal dynamics, such as classical and jazz. This architecture contributes to advancing AI-driven music analysis with applications in recommendation systems, music production, and automation across the music industry.

12 June 2026
Verified
Book Chapter
Closed access
GVG. Vennila
G. Vennila
DKDinesh Kumar K.
Dinesh Kumar K.
SAS. Anitha
S. Anitha
JRJariya Begum A. R.
Jariya Begum A. R.
The generation of diagnostic reports from chest X-ray images is a complex task that requires both accurate medical image interpretation and clear clinical language. Manual reporting is time-consuming, expertise-intensive, and susceptible to fatigue. Although deep learning–based systems have been developed to support this process, most traditional approaches treat image analysis and report generation as separate tasks, leading to inconsistencies and loss of critical details. Recent vision–language models offer a more unified solution by linking visual understanding with natural language generation. This chapter proposes an end-to-end framework that integrates a Swin Transformer image encoder with a Q-Former and a fine-tuned BioMedLM for producing accurate and clinically meaningful reports. Experiments on a curated chest X-ray dataset demonstrate improved pathology coverage, reduced errors, and enhanced report clarity, supporting reliable and efficient clinical decision-making.

30 June 2026
Verified
Book Chapter
Closed access
GVG. Vennila
G. Vennila
NDN. Duraimutharasan
N. Duraimutharasan
AVA. Vijay
A. Vijay
DKDinesh Kumar Kumar K.
Dinesh Kumar Kumar K.
The current development of Geospatial Artificial Intelligence (GeoAI) is making possible the location-aware decision support system which can improve precision in agriculture by analysing spatial data and making intelligent predictions. This chapter introduces a GeoAI-based multi-model agricultural intelligence system that comprises of Vision Transformers (ViT) to detect early plant diseases, Temporal Fusion Transformers (TFT) to predict the climate and Graph Neural Networks (GNN) to provide spatially aware crop recommendations. The GNN component captures the spatial relationships among soil properties, environmental factors, and crop yields to come up with adaptive outputs. A chatbot interface serves as a layer of homogeneous interaction, allowing the farmer to obtain real-time information with the lowest level of technical skills. The proposed GeoAI framework by combining spatial intelligence with multi-modal deep learning will be a step towards sustainable agriculture, lowered crop losses, and increased food security.

30 June 2026
Verified
Book Chapter
Closed access
GVG. Vennila
G. Vennila
PBPrabhu Shankar B.
Prabhu Shankar B.
KEKanniappan Elumalai
Kanniappan Elumalai
Plant diseases significantly impact negatively on agricultural productivity, resulting in huge financial losses and triggering a threat to the security of the world food supply. This study suggests a GeoAI-enabled hybrid deep learning framework that combines CoAtNet for reliable disease classification with Diffusion Models for image restoration in order to overcome these drawbacks. The suggested model was assessed using the PlantVillage dataset and artificially generated occluded images that simulate the real-world agricultural conditions. Based on the experiment findings, the Diffusion CoAtNet framework was able to perform better on degraded datasets compared to baseline CNN and Transformer-based models, with a 96.4% accuracy. Also, the system demonstrated good edge deployment performance, with a rate of approximately 2.4 frames per second (FPS) on Raspberry Pi hardware, and is suitable to real-time field tasks such as localized crop monitoring and disease diagnosis using a smartphone.