Target retrieval method based on group of randomized visual vocabularies and context semantic information
A visual dictionary and semantic information technology, applied in computer components, secure communication devices, character and pattern recognition, etc., can solve problems such as high computational complexity, achieve enhanced differentiation, reduce semantic gap, and solve computational complexity Effect
Patent Information
- Authority / Receiving Office
- CN · China
- Current Assignee / Owner
- Publication Date
- 2012-09-26
- Estimated Expiration
- Not applicable · inactive patent
Smart Images
Figure 1 Figure 2 Figure 3
Abstract
Description
technical field
[0001] The invention relates to a target retrieval method based on a randomized visual dictionary group and contextual semantic information. Background technique
[0002] In recent years, with the rapid development and application of computer vision, especially image local features (such as SIFT) and visual dictionary method (BoVW, Bag of Visual Words), the target retrieval technology has become more and more practical, and has been obtained in real-life products. widely used. For example, Tineye is a network-oriented near-duplicate image retrieval system, and Google Goggles allows users to use mobile phones to take pictures and retrieve information related to the objects contained in the pictures. The BoVW method is inspired by the word set method in the field of text retrieval. Due to its outstanding performance, the BoVW method has become the mainstream method in the field of target retrieval, but it also has some open problems. One is the low time effic...
Examples
Embodiment 1
[0071] Embodiment 1: This embodiment is based on the target retrieval method of randomized visual lexicon group and contextual semantic information. First, aiming at the low efficiency of traditional clustering algorithms and the problem of synonymous and ambiguous visual words, E 2 LSH clusters the local feature points of the training image database to generate a set of randomized visual dictionaries that support dynamic expansion; secondly, select the query image and use a rectangular frame to define the target area, and then extract the query image and the image database according to Lowe's method SIFT features and perform E on them 2 LSH mapping realizes the matching of feature points and visual words; then, on the basis of the language model, uses the rectangular frame area and image saliency detection to calculate the retrieval score of each visual word in the query image and obtain the target model containing the target context semantic information ;Finally, for the pro...
Embodiment 2
[0073] Embodiment two: see figure 2 , image 3 , Figure 4 , the target retrieval method based on the randomized visual dictionary group and contextual semantic information of this embodiment adopts the following steps to generate 2 LSH's randomized visual dictionary set:
[0074] For each hash function g i (i=1,...,L), use it to perform hash mapping on the SIFT points of the training image library, and the points with very close distances in the space will be stored in the same bucket of the hash table, with the center of each bucket represents a visual word, each function g i can generate a hash table, a visual dictionary. Then, L functions g 1 ,..., g L can generate a visual dictionary group, the process is as follows figure 2 shown.
[0075] Among them, the detailed process of single visual dictionary generation can be described as follows:
[0076] (1) SIFT feature extraction of the training image library. In this paper, Oxford5K, a commonly used database for...
Embodiment 3
[0103] Embodiment 3: The difference between this embodiment and Embodiment 2 is that the following steps are used to measure the similarity:
[0104] The similarity between the query image q and any image d in the image database can be measured by the query likelihood p(q|d), then:
[0105] p ( q | d ) = Π i = 1 M q p ( q i | d ) - - - ( 14 )
[0106] Transforming this into a risk minimization problem, that is, given a query image q, the risk function for returning an image d is defined as follows:
[0107]
[0108]
[0109] p(θ D |d)p(r|θ Q ,θ D )dθ Q dθ D ...