Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

General quality detection method of block average resolution combination without reference underwater images

An underwater image and image quality technology, applied in image enhancement, image analysis, image data processing and other directions, can solve the problems of blur, low contrast, light spots, etc., to achieve the effect of improving real-time performance, small calculation amount and high sensitivity

Active Publication Date: 2016-02-03
HUAIHAI INST OF TECH
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Most of the underwater images captured by the underwater image acquisition system have the following problems: limited viewing distance, low contrast, non-uniform illumination, blur, light spots, color loss and noise of various complex factors
This evaluation method will also be affected by noise, and it cannot be evaluated for untargeted underwater images and has poor visual consistency.

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
  • General quality detection method of block average resolution combination without reference underwater images
  • General quality detection method of block average resolution combination without reference underwater images
  • General quality detection method of block average resolution combination without reference underwater images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] Embodiment 1, a method for detecting the universality and quality of a block average definition combination without reference underwater images:

[0032] This method selects the global contrast and performs quality inspection based on local sharpness, and uses the result of image edge detection as the basis for judging whether there is a target; when detecting, the image with the target is divided into a series of image blocks, and the image blocks with measurement value The average value of sharpness evaluation and the weighted sum of global contrast and average brightness are used as a measure of image quality; when there is no target in the image, the stretchable range of contrast and average brightness are used as indicators to measure image quality; specifically The method steps are as follows:

[0033] for an underwater image I ,Assume I The image quality of Q , Q A larger value indicates that the image contains the target and is clearer, Q A smaller value in...

Embodiment 2

[0044] Embodiment 2, using the method of Embodiment 1 to conduct a detection experiment on any pair of underwater optical vision grayscale images:

[0045] Calculate the normalized contrast adjustment range, calculate the normalized cumulative histogram of the input grayscale image, and set I_Low to be the normalized gray level (0~ 1), I_High is the normalized gray level (0~1) corresponding to the first cumulative distribution reaching 0.99 in the histogram. T1=I_High-I_Low.

[0046] Calculate the average T of the normalized gray level of the image 2 . Assuming that the size of the image I is M×N, the normalized gray level average T 2 for:

[0047]

[0048] Carry out edge detection to image, adopt the algorithm of Sobel edge detection to obtain binary edge image in the embodiment of the present invention, but in the present invention, carry out edge detection to image and can adopt the method for any other single-pixel wide binary image edge detection.

[0049] Then, c...

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 method for universally detecting the quality of non-reference underwater images on the basis of a combination of block average definition. The method includes setting the quality Q of an underwater image I; setting a value of the quality Q as a weight of the contract and the average brightness of the underwater image if objects do not exist on the underwater image; determining that the objects exist on the image if the total amount of detected edges in the image is larger than a threshold value Edge_threshold<1>, selecting image blocks with detection significance if the quantities of edge pixels in the image blocks P are larger than a threshold value Edge_threshold<2>, and setting the value of the quality Q as the weighted sum of an average definition value, the contract and the average brightness of the image blocks. The objects are contained in the image and are clear along with increase of the value of the quality Q, and are not contained in the image or are blurred along with decrease of the value of the quality Q. The method has the advantages that quality detection values of all the underwater optical visual grayscale images can be provided by the method and are good in consistency with subjective judgment, and the problem that an existing method for evaluating the quality of partial underwater images only can be used for evaluating the underwater images of objects can be solved.

Description

technical field [0001] The invention belongs to the technical field of image processing and analysis, in particular to the processing and analysis of images taken in environments with absorption and scattering optical attenuation, such as underwater environments. The invention relates to the quality detection of underwater images, in particular to a generalized block-average underwater image quality detection method based on contrast and sharpness. Background technique [0002] The research of image quality objective evaluation algorithm is a very basic scientific problem in the field of image processing and image analysis. People hope to find a reliable and reasonable method for objective evaluation of image quality, which can approach the evaluation results of human visual system. In water, the speed of light is less affected by changes in temperature and salinity, the detection direction is good, and the positioning is more accurate. Target. Typical underwater visual i...

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): G06T7/00G06T5/00
Inventor 杨淼
Owner HUAIHAI INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products