Sri Phani Krishna Karri
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Sri Phani Krishna Karri has a total of 75 co-authors. Here are the top co-authors by number of shared publications:
Jyotirmoy Chatterjee
Indian Institute of Technology Kharagpur
13 shared publicationsDebdoot Sheet
Indian Institute of Technology Kharagpur
10 shared publicationsAjoy K. Ray
IIEST, Howrah, India
7 shared publications- Nagesh Bhattu Sristy
Department of Computer Science and Engineering, National Institute of Technology Andhra Pradesh, Tadepalligudem, India
6 shared publications Debjani Chakraborty
IIT Kharagpur, Kharagpur, India
5 shared publicationsHrushikesh Garud
4 shared publications
Latest Publications
- 23 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33 (10709908) , 2160-2164
Contrastive Dual-Prototypical Learning for Efficient Multi-Label Ophthalmic Diagnosis
Early screening for ophthalmic diseases is a public health imperative for preventing irreversible vision loss. However, real-world fundus datasets often exhibit multi-label comorbidities and long-tailed distributions, inducing diagnostic bias in conventional deep learning models. This letter proposes the Contrastive Dual-Prototypical Learning (CDPL) framework to address these challenges. CDPL utilizes a two-stage curriculum: Self-Supervised Learning (SSL) pre-training to extract invariant features, followed by orthogonality-regularized dual-prototypical fine-tuning that enforces angular separation to decouple overlapping pathological signals. Evaluated on the ODIR dataset using a ResNet-18 backbone, CDPL achieves a 9.7% Macro-F1 improvement over standard benchmarks. Qualitative t-SNE and Grad-CAM analyses confirm the structural refinement of the metric space and superior sensitivity to rare tail-class pathologies. The framework is designed for extensibility to multi-label medical imaging domains characterised by class imbalance, with theoretical generalizability supported by the dataset-agnostic construction of $\mathcal {L}_{\rm{ortho}}$.
- 23 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33 (10709908) , 2160-2164
Contrastive Dual-Prototypical Learning for Efficient Multi-Label Ophthalmic Diagnosis
Early screening for ophthalmic diseases is a public health imperative for preventing irreversible vision loss. However, real-world fundus datasets often exhibit multi-label comorbidities and long-tailed distributions, inducing diagnostic bias in conventional deep learning models. This letter proposes the Contrastive Dual-Prototypical Learning (CDPL) framework to address these challenges. CDPL utilizes a two-stage curriculum: Self-Supervised Learning (SSL) pre-training to extract invariant features, followed by orthogonality-regularized dual-prototypical fine-tuning that enforces angular separation to decouple overlapping pathological signals. Evaluated on the ODIR dataset using a ResNet-18 backbone, CDPL achieves a 9.7% Macro-F1 improvement over standard benchmarks. Qualitative t-SNE and Grad-CAM analyses confirm the structural refinement of the metric space and superior sensitivity to rare tail-class pathologies. The framework is designed for extensibility to multi-label medical imaging domains characterised by class imbalance, with theoretical generalizability supported by the dataset-agnostic construction of $\mathcal {L}_{\rm{ortho}}$.
- Open Access
- 3 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 7, 723-736
Deep Reinforcement Learning for Energy Management in Hybrid Electric Vehicles With Softmax Double-Actor Regularized Critics
Enhancing fuel efficiency in hybrid electric vehicles (HEVs) requires energy management strategies (EMSs) that can operate effectively under nonlinear powertrain dynamics and uncertain, time-varying driving conditions. This paper proposes a deep reinforcement learning (DRL)- based EMS using the double actors regularized critics softmax deep deterministic policy gradient (DARC SD3) algorithm, which integrates Boltzmann-softmax value estimation, a dual-actor architecture, and critic regularization to improve learning stability and value-estimation accuracy. Simulation results show that the proposed DARC SD3 achieves faster convergence, improved state-of-charge (SOC) regulation, and reduced value estimation bias compared with DDPG, TD3, and baseline SD3. Under the FTP-75 driving cycle, the proposed EMS attains 94.6% of the dynamic programming (DP) benchmark fuel economy, while reducing engine transients and smoothing battery power flow. Further evaluation on an unseen composite driving cycle confirms that the trained policy maintains consistent fuel economy and SOC control, demonstrating strong generalization capability across diverse driving conditions.
- 3 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 7, 723-736
Open AccessDeep Reinforcement Learning for Energy Management in Hybrid Electric Vehicles With Softmax Double-Actor Regularized Critics
Enhancing fuel efficiency in hybrid electric vehicles (HEVs) requires energy management strategies (EMSs) that can operate effectively under nonlinear powertrain dynamics and uncertain, time-varying driving conditions. This paper proposes a deep reinforcement learning (DRL)- based EMS using the double actors regularized critics softmax deep deterministic policy gradient (DARC SD3) algorithm, which integrates Boltzmann-softmax value estimation, a dual-actor architecture, and critic regularization to improve learning stability and value-estimation accuracy. Simulation results show that the proposed DARC SD3 achieves faster convergence, improved state-of-charge (SOC) regulation, and reduced value estimation bias compared with DDPG, TD3, and baseline SD3. Under the FTP-75 driving cycle, the proposed EMS attains 94.6% of the dynamic programming (DP) benchmark fuel economy, while reducing engine transients and smoothing battery power flow. Further evaluation on an unseen composite driving cycle confirms that the trained policy maintains consistent fuel economy and SOC control, demonstrating strong generalization capability across diverse driving conditions.
- 28 January 2026
- Published by Springer Nature in Journal of Electrical Engineering & Technology
- Vol. 21 (4) , 4039-4052
Attention-Driven YOLOv8 with Feature Fusion for Intelligent Insulator Defect Detection in Power Transmission Lines
- 28 January 2026
- Published by Springer Nature in Journal of Electrical Engineering & Technology
- Vol. 21 (4) , 4039-4052
Attention-Driven YOLOv8 with Feature Fusion for Intelligent Insulator Defect Detection in Power Transmission Lines
- Open Access
Cooperative Control and Energy Management for Autonomous Hybrid Electric Vehicles Using Machine Learning
The growing deployment of connected and autonomous vehicles (CAVs) requires coordinated control strategies that jointly address safety, mobility, and energy efficiency. This paper presents a novel two-stage cooperative control framework for autonomous hybrid electric vehicle (HEV) platoons based on machine learning. In the first stage, a metric learning-based distributed model predictive control (ML-DMPC) strategy is proposed to enable cooperative longitudinal control among heterogeneous vehicles, explicitly incorporating inter-vehicle interactions to improve speed tracking, ride comfort, and platoon-level energy efficiency. In the second stage, a multi-agent twin-delayed deep deterministic policy gradient (MATD3) algorithm is developed for real-time energy management, achieving an optimal power split between the engine and battery while reducing Q-value overestimation and accelerating learning convergence. Simulation results across multiple standard driving cycles demonstrate that the proposed framework outperforms conventional distributed model predictive control (DMPC) and multi-agent deep deterministic policy gradient (MADDPG)-based methods in fuel economy, stability, and convergence speed, while maintaining battery state of charge (SOC) within safe limits. To facilitate future experimental validation, a dSPACE-based hardware-in-the-loop (HIL) architecture is designed to enable real-time deployment and testing of the proposed control framework.
Open AccessCooperative Control and Energy Management for Autonomous Hybrid Electric Vehicles Using Machine Learning
The growing deployment of connected and autonomous vehicles (CAVs) requires coordinated control strategies that jointly address safety, mobility, and energy efficiency. This paper presents a novel two-stage cooperative control framework for autonomous hybrid electric vehicle (HEV) platoons based on machine learning. In the first stage, a metric learning-based distributed model predictive control (ML-DMPC) strategy is proposed to enable cooperative longitudinal control among heterogeneous vehicles, explicitly incorporating inter-vehicle interactions to improve speed tracking, ride comfort, and platoon-level energy efficiency. In the second stage, a multi-agent twin-delayed deep deterministic policy gradient (MATD3) algorithm is developed for real-time energy management, achieving an optimal power split between the engine and battery while reducing Q-value overestimation and accelerating learning convergence. Simulation results across multiple standard driving cycles demonstrate that the proposed framework outperforms conventional distributed model predictive control (DMPC) and multi-agent deep deterministic policy gradient (MADDPG)-based methods in fuel economy, stability, and convergence speed, while maintaining battery state of charge (SOC) within safe limits. To facilitate future experimental validation, a dSPACE-based hardware-in-the-loop (HIL) architecture is designed to enable real-time deployment and testing of the proposed control framework.
- Open Access
- 26 November 2025
- Published by Springer Nature in Research Square
Multi-Scale Encoder-Only Architectures for Enhanced Scene and Handwritten Text Recognition
Accurate and efficient recognition of scene and handwritten text is crucial for document analysis and computer vision applications. Traditional architectures often rely on encoder-decoder frameworks, which, despite their strong performance, suffer from high computational costs and inference latency due to the decoder's autoregressive nature. This work introduces encoder-only transformer architectures that eliminate the need for a decoder, enabling faster and more parallelizable computation. We propose four families of encoder-only models—PVTSTR, Twins-PCPVTSTR, Twins-SVTSTR, and VANSTR—by adapting state-of-the-art vision backbones to the domain of Scene Text Recognition (STR) and Handwritten Text Recognition (HTR). These models utilize progressive shrinking strategies for multi-scale feature extraction, allowing them to efficiently capture both fine-grained character-level features and global word-level semantics. Our models achieve accuracy improvements of up to 4.8\%, 0.5\%, and 3.6\% over baseline ViTSTR models, with specific variants such as VANSTR-b1 and Twins-SVTSTR-Small providing competitive accuracy while maintaining lower computational costs and faster inference speeds. Here we show that our encoder-only architectures significantly enhance text recognition performance, demonstrating their potential for a wide range of practical applications. The code of the proposed work is available in the repository doi.org/10.5281/zenodo. 15791181.
- 26 November 2025
- Published by Springer Nature in Research Square
Open AccessMulti-Scale Encoder-Only Architectures for Enhanced Scene and Handwritten Text Recognition
Accurate and efficient recognition of scene and handwritten text is crucial for document analysis and computer vision applications. Traditional architectures often rely on encoder-decoder frameworks, which, despite their strong performance, suffer from high computational costs and inference latency due to the decoder's autoregressive nature. This work introduces encoder-only transformer architectures that eliminate the need for a decoder, enabling faster and more parallelizable computation. We propose four families of encoder-only models—PVTSTR, Twins-PCPVTSTR, Twins-SVTSTR, and VANSTR—by adapting state-of-the-art vision backbones to the domain of Scene Text Recognition (STR) and Handwritten Text Recognition (HTR). These models utilize progressive shrinking strategies for multi-scale feature extraction, allowing them to efficiently capture both fine-grained character-level features and global word-level semantics. Our models achieve accuracy improvements of up to 4.8\%, 0.5\%, and 3.6\% over baseline ViTSTR models, with specific variants such as VANSTR-b1 and Twins-SVTSTR-Small providing competitive accuracy while maintaining lower computational costs and faster inference speeds. Here we show that our encoder-only architectures significantly enhance text recognition performance, demonstrating their potential for a wide range of practical applications. The code of the proposed work is available in the repository doi.org/10.5281/zenodo. 15791181.
- Open Access
- 1 October 2025
- Published by Springer Nature in Research Square
Scene Text Recognition via Alternating Hierarchical-Global Attention in Encoder-Only Transformers
Scene Text Recognition (STR) plays a pivotal role in various real-world applications, such as autonomous driving, document analysis, and assistive technologies, by enabling machines to extract and interpret text from natural images. Despite significant advancements in deep learning, existing transformer-based STR models, often suffer from high computational costs, large memory requirements, and slow inference due to the autoregressive nature of decoder-based architectures. This paper proposes two model variants, FasterViTSTR and DualFasterViTSTR, to address these challenges. FasterViTSTR is built upon the FasterViT architecture, modified to an encoder-only design for efficient STR. FasterViTSTR utilizes hierarchical attention to capture both local and global features, enhancing the model's ability to process variable-length sequences and complex spatial layouts. DualFasterViTSTR introduces a novel alternating hybrid attention mechanism, alternating between hierarchical and global attention layers, not only enhances performance but also improves efficiency. The proposed models are evaluated on multiple STR benchmark datasets, where the FasterViTSTR V2 variant achieves up to a 0.9% improvement in word accuracy over existing baselines, while the DualFasterViTSTR V2 variant delivers up to a 2.11% improvement, along with reduced inference time, FLOPs, and parameter count compared to FasterViTSTR. The implementation is made available in the repository : github.com/ShashNagendra/STR-via-Alternating-Hierarchical-Global-Attention.
- 1 October 2025
- Published by Springer Nature in Research Square
Open AccessScene Text Recognition via Alternating Hierarchical-Global Attention in Encoder-Only Transformers
Scene Text Recognition (STR) plays a pivotal role in various real-world applications, such as autonomous driving, document analysis, and assistive technologies, by enabling machines to extract and interpret text from natural images. Despite significant advancements in deep learning, existing transformer-based STR models, often suffer from high computational costs, large memory requirements, and slow inference due to the autoregressive nature of decoder-based architectures. This paper proposes two model variants, FasterViTSTR and DualFasterViTSTR, to address these challenges. FasterViTSTR is built upon the FasterViT architecture, modified to an encoder-only design for efficient STR. FasterViTSTR utilizes hierarchical attention to capture both local and global features, enhancing the model's ability to process variable-length sequences and complex spatial layouts. DualFasterViTSTR introduces a novel alternating hybrid attention mechanism, alternating between hierarchical and global attention layers, not only enhances performance but also improves efficiency. The proposed models are evaluated on multiple STR benchmark datasets, where the FasterViTSTR V2 variant achieves up to a 0.9% improvement in word accuracy over existing baselines, while the DualFasterViTSTR V2 variant delivers up to a 2.11% improvement, along with reduced inference time, FLOPs, and parameter count compared to FasterViTSTR. The implementation is made available in the repository : github.com/ShashNagendra/STR-via-Alternating-Hierarchical-Global-Attention.
- 4 September 2025
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 22 (5) , 1-14
Accurate insulator defect detection in power transmission lines using semi-supervised hybrid DETR with advanced loss methods
- 4 September 2025
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 22 (5) , 1-14
Accurate insulator defect detection in power transmission lines using semi-supervised hybrid DETR with advanced loss methods
- 25 August 2025
- Published by Elsevier in Measurement
- Vol. 257
Lightweight object detection for surface defect and health monitoring in renewable energy assets
- 25 August 2025
- Published by Elsevier in Measurement
- Vol. 257
Lightweight object detection for surface defect and health monitoring in renewable energy assets
- 18 June 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 293
Lightweight recurrence-free handwritten text recognition
- 18 June 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 293
Lightweight recurrence-free handwritten text recognition
- 30 November 2024
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 22 (1) , 1-15
Syflo: augmenting yolo for real-time health monitoring of electric assets in power transmission lines
- 30 November 2024
- Published by Springer Nature in Journal of Real-Time Image Processing
- Vol. 22 (1) , 1-15
Syflo: augmenting yolo for real-time health monitoring of electric assets in power transmission lines