Unlock instant, AI-driven research and patent intelligence for your innovation.

Method, apparatus, device and medium for generating super-resolution images based on feature separation

A high-resolution image and super-resolution technology, applied in the field of super-resolution image generation based on feature separation, can solve the problems of high computing power, expensive computing cost, and difficult network deployment, so as to reduce the number of feature channels and realize lightweight , reducing the effect of the parameter

Active Publication Date: 2022-08-02
中科天网(广东)科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] For the generation of super-resolution images, the key issue is how to enrich the details of low-resolution images in order to obtain more detailed information; on the other hand, although high computing power can obtain high-performance network models, high High-precision image restoration, but high computing power brings expensive computing costs and difficult network deployment

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
  • Method, apparatus, device and medium for generating super-resolution images based on feature separation
  • Method, apparatus, device and medium for generating super-resolution images based on feature separation
  • Method, apparatus, device and medium for generating super-resolution images based on feature separation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0081] This embodiment is based on the TensorFlow framework and the Pycharm development environment. The TensorFlow framework is a development framework based on the python language, which can easily and quickly build a reasonable deep learning network, and has good cross-platform interaction capabilities; it can provide interfaces for many encapsulation functions and various image processing functions in the deep learning framework. Including OpenCV-related image processing functions; at the same time, the GPU can be used to train and verify the model, which improves the calculation efficiency.

[0082] The Pycharm development environment (IDE) under Windows platform or Linux platform is one of the first choices for deep learning network design and development. Pycharm provides customers with new templates, design tools, and testing and debugging tools, and can provide customers with an interface to directly call remote servers.

[0083] like figure 1 As shown, this embodim...

Embodiment 2

[0143] like Image 6 As shown, this embodiment provides a feature separation-based super-resolution image generation device, the device includes an image acquisition module 601, a first feature extraction module 602, a second feature extraction module 603, a feature stacking module 604, and an optimization module. 605 and image reconstruction module 606, where:

[0144] An image acquisition module 601, configured to acquire a training data set; wherein, the image pairs in the training data set include low-resolution images and corresponding clear images;

[0145] The first feature extraction module 602 is configured to perform feature extraction on the low-resolution image by utilizing the feature extraction sub-network in the network model to obtain image features;

[0146] The second feature extraction module 603 is configured to perform deep feature extraction on the image features by utilizing the feature separation and recombination sub-network in the network model to ob...

Embodiment 3

[0152] This embodiment provides a computer device, and the computer device can be a computer, such as Figure 7 As shown, a processor 702, memory, input device 703, display 704, and network interface 705, which are connected by a system bus 701 for providing computing and control capabilities, the memory includes a non-volatile storage medium 706 and internal memory 707, the non-volatile storage medium 706 stores an operating system, a computer program and a database, the internal memory 707 provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium, and the processor 702 executes the memory stored in the memory. When computer program, realize the super-resolution image generation method of above-mentioned embodiment 1, as follows:

[0153] Obtain a training data set; wherein, the image pairs in the training data set include low-resolution images and corresponding clear images;

[0154] Use the feature extract...

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

No PUM Login to View More

Abstract

The invention discloses a super-resolution image generation method, device, computer equipment and storage medium based on feature separation. The method includes: acquiring a training data set; using a feature extraction sub-network in a network model to extract low-resolution images in the training data set Image feature extraction to obtain image features; use the feature separation and recombination sub-network in the network model to perform deep feature extraction on image features to obtain deep features; use the image reconstruction sub-network in the network model to characterize image features and deep features Superimpose to obtain a high-resolution image; optimize the loss function of the network model according to the clear image corresponding to the high-resolution image and the low-resolution image in the training dataset to obtain a trained network model; input the image to be tested into the trained network Models that generate super-resolution images. By utilizing the constructed network model, the present invention can obtain super-resolution images with more details, which lays a foundation for subsequent image processing and application.

Description

technical field [0001] The present invention relates to the technical field of deep learning applications, in particular to a method, device, computer equipment and storage medium for generating super-resolution images based on feature separation. Background technique [0002] In recent years, video surveillance has become popular in large and medium-sized cities across the country, and has been widely used in the construction of social security prevention and control systems. However, inevitably, affected by the resolution of the monitoring picture, some areas in the picture will be difficult to distinguish because the resolution is too low. In addition, in daily life, people also have high pixel requirements for the quality of the image: it is hoped that the details in the image can be clearly seen, so that the overall perception of the image is good. Therefore, it is necessary to perform super-resolution generation on low-resolution surveillance images to obtain images w...

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
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T3/40G06V10/40G06V10/82G06N3/04G06N3/08
CPCG06T3/4053G06T3/4046G06N3/08G06N3/045
Inventor 温峻峰张浪文李鑫杜海江江志伟谢巍杨晓峰
Owner 中科天网(广东)科技有限公司