Sketch-Based Image Retrieval System for Criminal Records

Sketch-Based Image Retrieval System for Criminal Records

Prof. Rupal More¹, Rajashri Puranik², Purva Dusane³ , Sejal Bhawar⁴,

Himanshu Sahu⁵

¹Assistant professor,  ^{2 3 4 5} Student

^{1 2 3 4  5} Department  of computer Engineering,

^{1 2 3 4 5}  Sandip Institute of  Technology and Research Center, Nashik,India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - An Overview of the facial recognition system using computer vision algorithms. The algorithm mainly uses the dlib library for finding the faces, estimating face landmarks, and Deep convolutional networks for pattern matching. Sketch-based image retrieval (SBIR) is a system where the user provides a sketch as input to the system to retrieve images relevant to the given sketch. Content-based image retrieval (CBIR) is widely used for retrieving images from substantial image databases. However, users are not satisfied with traditional information retrieval techniques. Therefore, the paper presents a simple and effective deep learning framework based on convolutional neural networks (CNN) for fast image retrieval, consisting of feature extraction and classification. From multiple studies on various CBIR tasks using image databases, we have yielded promising results that reveal important insights for improving the performance of CBIR.

Key Words: Deep Learning, Face recognition, DCNN, Feature extraction, Image retrieval