2020/04/22 by Manish Bhattarai, Diane Oyen, Bhattarai, Manish +7
Computer Science · Engineering · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image Retrieval and Classification Techniques #Robotics and Sensor-Based Localization
paper · pdf · doi:10.48550/arxiv.2004.10780
openalex publication_date 2020/04/22 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
Resolution of the complex problem of image retrieval for diagram images has\nyet to be reached. Deep learning methods continue to excel in the fields of\nobject detection and image classification applied to natural imagery. However,\nthe application of such methodologies applied to binary imagery remains limited\ndue to lack of crucial features such as textures,color and intensity\ninformation. This paper presents a deep learning based method for image-based\nsearch for binary patent images by taking advantage of existing large natural\nimage repositories for image search and sketch-based methods (Sketches are not\nidentical to diagrams, but they do share some characteristics; for example,\nboth imagery types are gray scale (binary), composed of contours, and are\nlacking in texture).\n We begin by using deep learning to generate sketches from natural images for\nimage retrieval and then train a second deep learning model on the sketches. We\nthen use our small set of manually labeled patent diagram images via transfer\nlearning to adapt the image search from sketches of natural images to diagrams.\nOur experiment results show the effectiveness of deep learning with transfer\nlearning for detecting near-identical copies in patent images and querying\nsimilar images based on content.\n