K. Devipriya
K. Devipriya has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
Budde Venkata Satya Tejesh
SRKR Engineering College,Dept. of Computer Science and Design,Bhimavaram,Andhra Pradesh,India,534204
1 shared publication- Chanumolu Kiran Kumar
Department of Computer Science and Engineering, Koneru Lakshmaiah Educational Foundation, Deemed to be University,Green Fields, Vaddeswaram, Andhra Pradesh, 522237, India
1 shared publication - Chiranjibe Jana
Saveetha School of Engineering, Saveetha Institute of Medical And Technical Sciences (SIMATS), Chennai, India
1 shared publication Kamparapu V V Satya Trinadh Naidu
SRKR Engineering College,Dept. of Computer Science and Information Technology,Bhimavaram,Andhra Pradesh,India,534204
1 shared publication- Mehdi Tlija
Industrial Engineering Department, College of Engineering, King Saud University, Riyadh, Saudi Arabia
1 shared publication Moturi Sandilya
SRKR Engineering College,Dept. of Computer Science and Design,Bhimavaram,Andhra Pradesh,India,534204
1 shared publication
Latest Publications
- 18 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
A Lightweight Smart Multimodal Document Question-Answering System with Embedding-Based Retrieval
The increase in the rate of digital resources in educational and organizational settings has challenged effective information retrieval. Traditional searches by use of key-word may result in the wrong or incomplete searches, due to lack of semantic meaning. The given paper provides a lightweight smart multimodal document question-answering (QA) system that utilizes embedding-based retrieval and large language models (LLMs) to answer questions correctly and in a context-sensitive manner. The documents uploaded by a faculty or an organization are embedded and stored in a vector database (ChromaDB) and can be searched by similarity effectively. The query of users is processed into embeddings, and the appropriate document snippets are retrieved and synthesized into sensible responses with the help of LLMs. The system also uses Gemini and Sarvam APIs to use text-to-speech features and speech-to-text functions to provide responses in both text and audio formats to promote increased accessibility. Being lightweight, the system can be deployed in low-resource environments and therefore is feasible in the case of educational institutions and organizations that have a small computational infrastructure. The solution proposed illustrates a high potential of managing knowledge which has the ability to save manual work and enhances the efficiency and inclusiveness in the access of information.
- 18 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
A Lightweight Smart Multimodal Document Question-Answering System with Embedding-Based Retrieval
The increase in the rate of digital resources in educational and organizational settings has challenged effective information retrieval. Traditional searches by use of key-word may result in the wrong or incomplete searches, due to lack of semantic meaning. The given paper provides a lightweight smart multimodal document question-answering (QA) system that utilizes embedding-based retrieval and large language models (LLMs) to answer questions correctly and in a context-sensitive manner. The documents uploaded by a faculty or an organization are embedded and stored in a vector database (ChromaDB) and can be searched by similarity effectively. The query of users is processed into embeddings, and the appropriate document snippets are retrieved and synthesized into sensible responses with the help of LLMs. The system also uses Gemini and Sarvam APIs to use text-to-speech features and speech-to-text functions to provide responses in both text and audio formats to promote increased accessibility. Being lightweight, the system can be deployed in low-resource environments and therefore is feasible in the case of educational institutions and organizations that have a small computational infrastructure. The solution proposed illustrates a high potential of managing knowledge which has the ability to save manual work and enhances the efficiency and inclusiveness in the access of information.
- Open Access
- 1 February 2025
- Published by AIP Publishing in AIP Advances
- Vol. 15 (2)
Automated micro-plastic detection and classification using deep convolution neural network pre-trained models and transfer learning
Micro-plastics, which are plastic particles less than 5 mm in size, pose significant environmental threats due to their persistence and potential toxicity to marine life and humans. This work compares the effectiveness of several convolutional neural network (CNN) designs, including MobileNetV3Large, ResNet50V2, ResNet101V2, and EfficientNetB7, in identifying and categorizing microplastics into three groups: beads, fibers, and fragments. We evaluate the models using precision, recall, and F1-score criteria. The outcomes indicate that while all models perform well in identifying fiber microplastics, achieving high recall and moderate precision, they struggle significantly with bead and fragment categories. EfficientNetB7 and MobileNetV3Large exhibited the highest performance for fiber detection but failed to detect bead and fragment microplastics. The findings highlight the need for further research to enhance the classification accuracy for bead and fragment micro-plastics, suggesting that future work should focus on addressing class imbalance, utilizing advanced techniques such as transfer learning, and incorporating domain-specific knowledge to improve feature discrimination. This work provides the path for more efficient environmental monitoring systems by offering insightful information about the advantages and disadvantages of the CNN architectures now in use for the detection of microplastics.
- 1 February 2025
- Published by AIP Publishing in AIP Advances
- Vol. 15 (2)
Open AccessAutomated micro-plastic detection and classification using deep convolution neural network pre-trained models and transfer learning
Micro-plastics, which are plastic particles less than 5 mm in size, pose significant environmental threats due to their persistence and potential toxicity to marine life and humans. This work compares the effectiveness of several convolutional neural network (CNN) designs, including MobileNetV3Large, ResNet50V2, ResNet101V2, and EfficientNetB7, in identifying and categorizing microplastics into three groups: beads, fibers, and fragments. We evaluate the models using precision, recall, and F1-score criteria. The outcomes indicate that while all models perform well in identifying fiber microplastics, achieving high recall and moderate precision, they struggle significantly with bead and fragment categories. EfficientNetB7 and MobileNetV3Large exhibited the highest performance for fiber detection but failed to detect bead and fragment microplastics. The findings highlight the need for further research to enhance the classification accuracy for bead and fragment micro-plastics, suggesting that future work should focus on addressing class imbalance, utilizing advanced techniques such as transfer learning, and incorporating domain-specific knowledge to improve feature discrimination. This work provides the path for more efficient environmental monitoring systems by offering insightful information about the advantages and disadvantages of the CNN architectures now in use for the detection of microplastics.