Noise-possessing movement fuzzy image restoration method based on radial basis nerve network

A neural network-based technology for motion blurred images, applied in image enhancement, image data processing, instruments, etc., can solve problems such as high computational complexity, unsatisfactory restoration effect, and inability to realize automatic identification

Inactive Publication Date: 2007-11-28
ZHEJIANG NORMAL UNIVERSITY
View PDF0 Cites 32 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the shortcomings of the existing noisy motion blurred image restoration methods that cannot realize automatic identification, high computational complexity, and unsatisfactory restoration 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
  • Noise-possessing movement fuzzy image restoration method based on radial basis nerve network
  • Noise-possessing movement fuzzy image restoration method based on radial basis nerve network
  • Noise-possessing movement fuzzy image restoration method based on radial basis nerve network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0078] With reference to Fig. 1~Fig. 6, Fig. 8, a kind of noisy motion-blurred image restoration method based on radial basis neural network, this restoration method comprises the following steps:

[0079] (1), define the noisy motion blurred image as y(m, n), first use the two-dimensional median filter to generate a corresponding low-pass filtered smooth image for the noisy motion blurred image y(m, n), and calculate The formula is:

[0080] s ( m , n ) = 1 ( 2 K + 1 ) 2 Σ i = - K K Σ j = - K K ...

Embodiment 2

[0126] Referring to Figures 1 to 8, in this embodiment, when the local motion blurred objects are extracted, it is found that the newly generated images that only contain the range of blurred objects clearly show the characteristics of ordinary motion blurred images, which can be completely obtained by The automatic identification image restoration method proposed in Embodiment 1 can effectively restore it.

[0127] For the restoration of a single local motion blurred image, the key problem is to extract the motion blurred object when the single frame image lacks the reference information of the relevant sequence frame. For example, taking a single-frame multi-lane road condition monitoring system video image as an example, as shown in Figure 7, it can be considered to use prior knowledge to simplify the problem. The objects to be extracted in the application are mainly high-speed moving cars in each lane, and the car’s The shape is close to a rhomboidal rectangle. Accordingly...

Embodiment 3

[0135] Referring to Figures 1 to 8, another target object extraction method is used in this embodiment. The algorithm in embodiment 2 has a small amount of computation and is fast and convenient to implement. For some single images with relatively close grayscale values, interception and segmentation with grayscale thresholds may result in large regional errors. Consider using Algorithm 5 to achieve target object extraction for such fuzzy images:

[0136] Algorithm 5. Local motion blur object extraction algorithm.

[0137] Step1. Comprehensively use the Prewitt operator and the Canny operator logic and operation on the motion blur image to perform edge detection (high accuracy, and weak edges can be detected);

[0138] Step2. Use Radon transform on the binary edge image to detect all lengths greater than L min The line segments, save their starting point, end point and the angle between the horizontal direction (can overcome the shortcomings of the traditional Hough transform...

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 noise motion blurred picture recovery method based on radial neural net, which comprises the following steps: (1) making the noise motion blurred picture y(m, n) generate the corresponding smoothing picture s(m, n) after low-pass filtering with the two-dimensional median filter; (2) calculating the inaccuracy picture e(m, n)=y(m, n)-s(m, n) between the noise motion blurred picture and the smoothing picture; (3) detecting the edge with Canny operator in order to estimate the gradient f' (m, n) of y(m, n); (4) calculating regular parameter lambda (m, n) of every picture element according to the size of f' (m, n) and generating interpolation picture f lambda (m, n) by radial neural net RBFN according to the inaccuracy picture e(m, n); (5) acquiring the silent motion blurred picture f(m, n) by superimposing the interpolation picture f lambda (m, n) to the smoothing picture s(m, n) after low-pass filtering; (6) acquiring the width HL and the height VL of two-dimensional blurred picture by identifying the motion blurred direction and the motion blurred length of the motion blurred picture automatically and acquiring the recovery picture with the picture recovery method. The invention provides the automatic identification, the low complicated calculation and the good recovery effect.

Description

technical field [0001] The invention relates to a method for restoring noisy motion blurred images. Background technique [0002] The research and application fields of noisy and blurred image restoration are very extensive, such as high-speed motion blur restoration of illegal or accident license plate, sudden suspect capture defocus or motion blur restoration, crime scene trace blur restoration, video surveillance specific frame blur restoration, etc. Generally, the imaging degradation is caused by defocus blur during the imaging process, relatively high-speed motion of imaging equipment and objects, inherent defects of equipment, shooting shake, and external noise interference. Among them, it is especially difficult to clear and restore a single noisy blurred image with local non-uniform motion. The reason is that the blurring of such images is complicated, the image damage is large, and there is no front and rear related sequence frames for reference...

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
IPC IPC(8): G06T5/00
Inventor 朱信忠赵建民徐慧英章琳
Owner ZHEJIANG NORMAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap