Fast image copy detection approach based on local fingerprint defined visual words

Design of Complicated Duplicate Image Representation Approach Based on Descriptor Learning

“…So visual word based on the unsupervised learning is more discriminative than based on Kmeans clustering definition. The related experiment results also prove this conclusion [5] . However, the unsupervised learning process in the generation of a visual dictionary does not consider the semantic information.…”

“…In this section, compares this algorithm with the algorithm in literature [6] and [5]. We can see that this algorithm has the best retrieval results from Figure 5.…”

“…The average retrieval precision is only 68.5%. While this algorithm and literature [5] is extracted image local descriptors. It make the binary encoding set of local descriptor on contains more information.…”

“…In order to solve the above problems, literature [5] use unsupervised learning of local descriptor quantization image into binary code to improve the image distinguishes ability. Visual dictionary as a result of unsupervised learning can be great.…”

Effective and Efficient Image Copy Detection with Resistance to Arbitrary Rotation

“…To address the problem of the robustness to rotation and some other geometric transformations, recently, some local feature-based image copy detection methods have been proposed [17]- [22]. They have achieved desirable perfor- mances for image copy detection because of the good robustness of local features.…”