Super-resolution reconstruction method for single image from rough to fine
A technology of super-resolution reconstruction and single image, which is applied in the field of computer vision, can solve the problem of low utilization rate of high-resolution spatial information, and achieve the effect of increasing the receptive field and comprehensive image context information
Active Publication Date: 2020-05-15
DALIAN UNIV OF TECH
View PDF4 Cites 9 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0007] Aiming at the technical problems that it is difficult to effectively extract and utilize image context feature information and low utilization rate of high-resolution spatial information in the process of image super-resolution reconstruction, the present invention designs a single image super-resolution reconstruction algorithm based on deep learning technology, which can Given a low-resolution image, generate a high-resolution image with rich details and clear textures
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0043] The present invention will be described in further detail below in conjunction with specific embodiments, but the present invention is not limited to specific embodiments.
[0044] A single image super-resolution reconstruction method based on two stages from coarse to fine, including network model training and model evaluation
[0045] (1) Training network model
[0046] Firstly, the images in DIV2K are down-sampled by bicubic interpolation to obtain low-resolution images, and the low-resolution images are rotated and flipped to obtain training data. At the same time according to Figure 4 to build the neural network model; then the training data is multi-threaded and batched into the network model to be trained, and the error between the high-resolution image reconstructed by the neural network and the true value is calculated according to the formula (4); finally, according to the reverse The propagation method uses the gradient descent optimizer ADAM to iterativel...
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 belongs to the technical field of computer vision, and provides a super-resolution reconstruction method for a single image from rough to fine, and the whole network comprises two stages: a multi-context extraction stage and a reconstruction enhancement stage, wherein the multi-context extraction stage is used for extracting image context feature information of a low-resolution space, and the reconstruction enhancement stage is used for extracting feature information of a high-resolution space. The reconstructed high-resolution image is good in visual effect and excellent in image evaluation indexes such as peak signal-to-noise ratio and structural similarity. Meanwhile, the method is low in time cost and hardware requirements.
Description
technical field [0001] The invention belongs to the technical field of computer vision, in particular to a method for super-resolution reconstruction of a single image based on deep learning. Background technique [0002] Image resolution refers to the amount of information stored in an image, and is an important indicator to measure an image. High-resolution images contain more detailed texture information and have a stronger ability to express information. With the widespread use of electronic products with higher resolution, people's demand for high-resolution images has become greater and greater. Single image super-resolution reconstruction is to reconstruct a high-resolution image from a given low-resolution image. Image super-resolution reconstruction can be applied to many important fields, for example: in medical image MRI, CT, super-resolution reconstruction can better help doctors determine the detailed condition of the disease; in surveillance video, image supe...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06T3/40G06N3/04
CPCG06T3/4053G06T3/4046G06N3/045
Inventor 张吉庆杨鑫尹宝才魏小鹏张强
Owner DALIAN UNIV OF TECH