Dr. Praveen Chitti

EDUCATION

B.E. (E&C), M. Design & Embedded Systems), Ph. D.

RESEARCH INTERESTS

Image Processing, Cloud Computing

Scopus Publications

A Deep Learning and Machine Learning Approach for Image Classification of Tempered Images in Digital Forensic Analysis
Praveen Chitti, K. Prabhushetty, and Shridhar Allagi
The Science and Information Organization
—Multimedia images are the primary source of communication across social media and other websites. Multimedia security has gained the attention of modern researchers and has posed dynamic challenges such as image forensics, image tampering, and deep fakes. Malicious users tamper with the image embedding noise, leading to misinterpretation of the content. Identifying and authenticating the image by detecting the forgery operations performed on it is essential. In our proposed model, we detect the forged region using the machine learning model SVM in the first iteration and Convolution Neural Network in the second iteration with Discrete Cosine Transform (DCT) for feature extraction. The proposed model is tested with a Corel 10K dataset, and an average accuracy of 98% is obtained for all kinds of image operations, including scaling, rotation, and augmentation.


Message
Chenraj Roychand
IEEE



Message
IEEE



Secured CBIR with Anonimity Preserving for Images and Users in Cloud Environment
Praveen. Y. Chitti and K. Prabhushetty
IEEE
With advancement in imaging devices, the images are playing a vital role in everyday life. Multimedia image usage across various platforms are increased by major usage of communication technologies. The images in medical, automotive and various other domain carries the sensitive information in images. Handling images on large scale is expensive and computationally tedious. Major organization outsource the centralized image repository and other computations on image to cloud server. The cloud being the cheapest way of handling operations with images but lacks the concern of security and privacy. Example, the images of patients in hospitals and other personal images have to be encrypted before outsourcing to the cloud vendor, Content Based Image Retrieval (CBIR) then can be used for querying image. The Paper presents a novel method of outsourcing image repository without revealing information about images to cloud vendors by encrypting image dataset. The extracted feature vectors are used to classify the images are also encrypted, and a local hashing table is developed to improve the search efficiency. The generated feature vector values are secured using kNN. The experiment results have shown acceptable level of efficiency.