Abhrendu Bhattacharya
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Abhrendu Bhattacharya has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Paramita Sarkar
JIS University
6 shared publications - Amit R. Welekar
J. D. College of Engineering and Management
3 shared publications - Achyut Mitra
JIS University
2 shared publications - Rajesh Bose
JIS University
2 shared publications - Sayak Nayek
Department of Computer Science and Engineering, IS University, Kolkata, India
2 shared publications - Somnath Maity
Department of Computer Science and Engineering, IS University, Kolkata, India
2 shared publications
Latest Publications
- 2 September 2026
- Published by Springer Nature
- p. 1-10
A Systematic Review of Artificial Intelligence Techniques for Handwriting-Based Cognitive and Behavioral State Assessment
- 2 September 2026
- Published by Springer Nature
- p. 1-10
A Systematic Review of Artificial Intelligence Techniques for Handwriting-Based Cognitive and Behavioral State Assessment
- 28 August 2026
- Published by Springer Nature in Applied Fruit Science
- Vol. 68 (5) , 301
Hybrid Field-Programmable Gate Array-Optimized Ripeness Detection for Asymmetric Fruits
- 28 August 2026
- Published by Springer Nature in Applied Fruit Science
- Vol. 68 (5) , 301
Hybrid Field-Programmable Gate Array-Optimized Ripeness Detection for Asymmetric Fruits
- 2 August 2026
- Published by Springer Nature
- p. 247-258
CO2-to-Light: Smart Solutions for Using Highway Emissions for Sustainable Energy in Smart Cities
- 2 August 2026
- Published by Springer Nature
- p. 247-258
CO2-to-Light: Smart Solutions for Using Highway Emissions for Sustainable Energy in Smart Cities
- Open Access
- 31 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1) , 29040
Secure dual biometric image steganography via CNN-refined chaotic hybrid transform embedding
- 31 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1) , 29040
Open AccessSecure dual biometric image steganography via CNN-refined chaotic hybrid transform embedding
- Open Access
- 1 June 2026
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 15 (3)
A scalable low-cost internet of things-based electronic nose for identifying chemical ripening in fruits
Indiscriminate use of chemical agents like calcium carbide and ethephon for the ripening of fruits poses grave health hazards, emitting carcinogenic and neurotoxic compounds. Here we present a new, scalable, inexpensive, internet of things (IoT)-enabled electronic nose (e-nose) AI-Bot system for the detection of chemically ripened fruits. This would involve the development of a system that uses an MQ-3 gas sensor to quantify the ethanol content, as well as an MQ-135 gas sensor with an ESP32 microcontroller to quantify even further the amount of volatile organic compounds (VOCs) suggestive of artificial ripening. Flutter-based mobile application allows real-time monitoring, ripening classification using machine learning (ML) algorithms, and logging the historical data. A small sample was taken for inter-document feature literature mining, modelling sensor behaviors according to voltage dividers and gas concentration resistance laws for robust calibration and classification performance. Validation studies were performed on mango, banana, and papaya fruit in the laboratory environment. Total 75 samples (25 each of banana, mango, papaya across 3 trials) of fruit were tested. The implemented system achieved 95% for banana, 92% for mango, and 90% classification accuracy for papaya when cross validated.
- 1 June 2026
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 15 (3)
Open AccessA scalable low-cost internet of things-based electronic nose for identifying chemical ripening in fruits
Indiscriminate use of chemical agents like calcium carbide and ethephon for the ripening of fruits poses grave health hazards, emitting carcinogenic and neurotoxic compounds. Here we present a new, scalable, inexpensive, internet of things (IoT)-enabled electronic nose (e-nose) AI-Bot system for the detection of chemically ripened fruits. This would involve the development of a system that uses an MQ-3 gas sensor to quantify the ethanol content, as well as an MQ-135 gas sensor with an ESP32 microcontroller to quantify even further the amount of volatile organic compounds (VOCs) suggestive of artificial ripening. Flutter-based mobile application allows real-time monitoring, ripening classification using machine learning (ML) algorithms, and logging the historical data. A small sample was taken for inter-document feature literature mining, modelling sensor behaviors according to voltage dividers and gas concentration resistance laws for robust calibration and classification performance. Validation studies were performed on mango, banana, and papaya fruit in the laboratory environment. Total 75 samples (25 each of banana, mango, papaya across 3 trials) of fruit were tested. The implemented system achieved 95% for banana, 92% for mango, and 90% classification accuracy for papaya when cross validated.
- 14 May 2026
- Published by Taylor & Francis
- p. 18-30
Next-Gen Customer Intelligence
- Open Access
- 13 February 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Adaptive bit placement for dual biometric using signature and finger print for DWT-DCT picture steganography
Providing biometric data through multimedia has made it difficult to ensure secure communication and authentication in the digital age. This research showcases an advanced dual-biometric steganographic system that incorporates signature and fingerprint characteristics in a single cover image using Discrete Wavelet Transform (DWT) and Disputed Cosine Transform (DCT). An Adaptive Bit Placement (ABP) approach is proposed, which intelligently selects embedding positions by considering coefficient significance and biometric importance to balance imperceptibility, robustness, and embedding.To verify its efficacy, the system was compared to standard benchmark images and provided with live biometric data samples. The experimental outcomes demonstrate that the proposed ABP-based Dual Biometric Steganography (ABPDBS) is significantly superior to current methods. While the current methods show Peak Signal to Noise Ratio (PSNR) values between 34 and 39 dB and Structural Similarity Index Measurement (SSIM) scores between 0.90 and 0.96, and embedding capacities below 0.30 bpp, the proposed approach yields average PSNR improvements of 22–36% (47.31-50.76dB), mean value of SSIM improvements of 2–11% (0.9748 with maximum reach to 0.9995) and an embedding capacity increases of 16% (0.33 bpp). Proposed method also provides overall better integration performance in these new techniques than those previously described. These results imply that the system is highly imperceptible; resilient to typical signal processing attacks, and has greater payload strength. In general, this research presents a secure and efficient framework for dual-biometric embedding which can enhance multimodal authentication systems suitable for high-security scenarios.
- 13 February 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessAdaptive bit placement for dual biometric using signature and finger print for DWT-DCT picture steganography
Providing biometric data through multimedia has made it difficult to ensure secure communication and authentication in the digital age. This research showcases an advanced dual-biometric steganographic system that incorporates signature and fingerprint characteristics in a single cover image using Discrete Wavelet Transform (DWT) and Disputed Cosine Transform (DCT). An Adaptive Bit Placement (ABP) approach is proposed, which intelligently selects embedding positions by considering coefficient significance and biometric importance to balance imperceptibility, robustness, and embedding.To verify its efficacy, the system was compared to standard benchmark images and provided with live biometric data samples. The experimental outcomes demonstrate that the proposed ABP-based Dual Biometric Steganography (ABPDBS) is significantly superior to current methods. While the current methods show Peak Signal to Noise Ratio (PSNR) values between 34 and 39 dB and Structural Similarity Index Measurement (SSIM) scores between 0.90 and 0.96, and embedding capacities below 0.30 bpp, the proposed approach yields average PSNR improvements of 22–36% (47.31-50.76dB), mean value of SSIM improvements of 2–11% (0.9748 with maximum reach to 0.9995) and an embedding capacity increases of 16% (0.33 bpp). Proposed method also provides overall better integration performance in these new techniques than those previously described. These results imply that the system is highly imperceptible; resilient to typical signal processing attacks, and has greater payload strength. In general, this research presents a secure and efficient framework for dual-biometric embedding which can enhance multimodal authentication systems suitable for high-security scenarios.
- Open Access
- 25 January 2026
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 6 (1)
Harnessing ancient Vedic mathematics with Karatsuba multiplication for efficient and secure face recognition in surveillance
- 25 January 2026
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 6 (1)
Open AccessHarnessing ancient Vedic mathematics with Karatsuba multiplication for efficient and secure face recognition in surveillance
- Open Access
- 27 September 2025
- Published by Academic Publications in International Journal of Applied Mathematics
- Vol. 38 (4s) , 1401-1421
COVERLESS IMAGE STEGANOGRAPHY USING DEEP IMAGE SEMANTICS AND ROBUST HASHING
With the rapid growth of digital communication, ensuring that data is transmitted securely, imperceptible, and without any form of interference is now crucial. Due to the vulnerability of traditional steganographic techniques that involve pixel-level modification, they are increasingly vulnerable to steranalysis attacks and statistical weaknesses as well as content degradation, rendering them unsuitable for sensitive areas such as healthcare, biometrics, and digital forensics. Our research in this paper proposes a fresh approach to Coverless Image Steganography using Deep Image Semantics and Robust Hashing (CISDSRH), which eliminates the need for image manipulation entirely by extracting semantic-level features and applying firmly hash-based mapping. Extracting high-dimensional semantic vectors from an image set that can be visualized is the primary objective of the technique, which employs deep convolution neural networks. These vectors are then converted to stable binary hash codes by means of methods such as LSH and ITQ. The hash-friendly binary segments are used to encode the secret message, which is presented through a carefully chosen sequence of semantically matched images, resulting in secure; high-capacity embedding that preserves the original image intact. It is clear from extensive experiments on benchmark datasets that the proposed system has perfect perceptual integrity (PSNR = ∞, SSIM =99%) and zero-bit error (BER 0.0031) under ideal conditions.
- 27 September 2025
- Published by Academic Publications in International Journal of Applied Mathematics
- Vol. 38 (4s) , 1401-1421
Open AccessCOVERLESS IMAGE STEGANOGRAPHY USING DEEP IMAGE SEMANTICS AND ROBUST HASHING
With the rapid growth of digital communication, ensuring that data is transmitted securely, imperceptible, and without any form of interference is now crucial. Due to the vulnerability of traditional steganographic techniques that involve pixel-level modification, they are increasingly vulnerable to steranalysis attacks and statistical weaknesses as well as content degradation, rendering them unsuitable for sensitive areas such as healthcare, biometrics, and digital forensics. Our research in this paper proposes a fresh approach to Coverless Image Steganography using Deep Image Semantics and Robust Hashing (CISDSRH), which eliminates the need for image manipulation entirely by extracting semantic-level features and applying firmly hash-based mapping. Extracting high-dimensional semantic vectors from an image set that can be visualized is the primary objective of the technique, which employs deep convolution neural networks. These vectors are then converted to stable binary hash codes by means of methods such as LSH and ITQ. The hash-friendly binary segments are used to encode the secret message, which is presented through a carefully chosen sequence of semantically matched images, resulting in secure; high-capacity embedding that preserves the original image intact. It is clear from extensive experiments on benchmark datasets that the proposed system has perfect perceptual integrity (PSNR = ∞, SSIM =99%) and zero-bit error (BER 0.0031) under ideal conditions.