Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
An image retrieval and image technology, applied in the field of image processing, can solve problems such as insurmountable, slow image retrieval progress, poor image retrieval effect, etc.
Inactive Publication Date: 2017-06-23
BOCOM SMART INFORMATION TECH CO LTD
View PDF7 Cites 15 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0004] Therefore, the technical problem to be solved by the embodiments of the present invention is that the image retrieval method in the prior art completes the image retrieval of the single-task learning feature through a single attribute, and cannot learn the relationship between the various attributes of the image and the detailed features of the image. In the process, it is impossible to overcome the influence of climate, environment, and illumination, resulting in the slowdown of image retrieval progress and poor image retrieval effect
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0077] This embodiment provides an image retrieval method, such as figure 1 shown, including the following steps:
[0078] S1. Build a multi-task deep network structure model; the multi-task deep network structure model here is to use the deep neural network learning platform to realize the multi-task learning framework, so that there is room for multi-task deep learning;
[0079] As an implementation, the image retrieval method in this embodiment, step S1, includes:
[0080] S11. Determine the attribute category of the target image and the training image subset corresponding to the attribute category; in the process of retrieving images, because the image types are rich and colorful, it is necessary to divide multiple images in detail to retrieve the target image more easily. For example, in the image attribute classification of people, there is a certain correlation between wearing lipstick and wearing earrings. In the attribute classification of cars, red Faradays are more...
Embodiment 2
[0103] This embodiment provides an image retrieval device, including the following units:
[0104] A construction unit 51, configured to construct a multi-task deep network structure model;
[0105] Establishing unit 52, for establishing target image feature library;
[0106] The input unit 53 is used to input the feature subset of the image to be retrieved;
[0107] Calculation unit 54, used to calculate the similarity distance between the feature subset of the image to be retrieved and each image feature in the image feature library;
[0108] The acquisition unit 55 acquires the image with the smallest distance to the image to be retrieved after sorting the distances from small to large.
[0109] As an implementation, the image retrieval device in this embodiment, such as Image 6 As shown, building unit 51 includes:
[0110] The first determining module 511 is used to determine the attribute category of the target image and the training image subset corresponding to the a...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention discloses an image retrieval method and apparatus. The image retrieval method comprises the following steps of building a multi-task depth network structure model; establishing a target image feature library; inputting a feature subset of a to-be-retrieved image; calculating a similarity distance between the feature subset of the to-be-retrieved image and each image feature in the image feature library; and performing arrangement according to an arrangement order of distances from short to long to obtain an image with a minimum distance with the to-be-retrieved image. By building the depth network structure model for multi-task learning of the features of the target image, the image retrieval is finished; the precision of a plurality of attribute classifications can be jointly improved by fully utilizing a correlation of tasks; and a relationship of the attributes of the image and detail features of the image can be learnt, so that in the image retrieval process, the influence of climate, environment, illuminance and the like can be overcome.
Description
technical field [0001] The invention relates to the field of image processing, in particular to an image retrieval method and device. Background technique [0002] With the rapid development of computers and the Internet, multimedia image information is being rapidly generated and disseminated on the Internet, thereby enriching people's vision. Faced with various image information, mining the images that people need has become a concern. Therefore, the establishment of relevant image retrieval methods has become a hot spot in research and engineering practice. [0003] At present, the existing image retrieval methods generally use the deep learning of neural networks to directly apply a single attribute to complete the image retrieval of single-task learning features, because single-task learning features often only consider the amount of information of the task itself, and do not consider the relationship with other tasks. The connection between tasks, so the learning abil...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more