WO2022046016A1 - Variable clipping level calculation method for clahe algorithm - Google Patents

Variable clipping level calculation method for clahe algorithm Download PDF

Info

Publication number
WO2022046016A1
WO2022046016A1 PCT/TR2021/050869 TR2021050869W WO2022046016A1 WO 2022046016 A1 WO2022046016 A1 WO 2022046016A1 TR 2021050869 W TR2021050869 W TR 2021050869W WO 2022046016 A1 WO2022046016 A1 WO 2022046016A1
Authority
WO
WIPO (PCT)
Prior art keywords
histogram
window
pixel
clipping level
value
Prior art date
Application number
PCT/TR2021/050869
Other languages
French (fr)
Inventor
Muhammed Mehdi MAVİ
Original Assignee
Aselsan Elektroni̇k Sanayi̇ Ve Ti̇caret Anoni̇m Şi̇rketi̇
İstanbul Tekni̇k Üni̇versi̇tesi̇
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Aselsan Elektroni̇k Sanayi̇ Ve Ti̇caret Anoni̇m Şi̇rketi̇, İstanbul Tekni̇k Üni̇versi̇tesi̇ filed Critical Aselsan Elektroni̇k Sanayi̇ Ve Ti̇caret Anoni̇m Şi̇rketi̇
Priority to EP21862237.1A priority Critical patent/EP4091130A4/en
Publication of WO2022046016A1 publication Critical patent/WO2022046016A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/90—Dynamic range modification of images or parts thereof
    • G06T5/94—Dynamic range modification of images or parts thereof based on local image properties, e.g. for local contrast enhancement
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/40—Image enhancement or restoration using histogram techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20004—Adaptive image processing
    • G06T2207/20012—Locally adaptive

Definitions

  • the invention relates to a pixel-specific clipping level parameter calculation method for the real-time sliding window CLAHE (contrast-limited adaptive histogram equalization algorithm based on edge information around each pixel.
  • CLAHE contrast-limited adaptive histogram equalization algorithm based on edge information around each pixel.
  • a separate region histogram is calculated for all pixels in the image. Region histograms calculated for each pixel are clipped from a fixed clipping level, and values clipped from the region histogram are redistributed to the entire region histogram. Then, the output value of the relevant pixel is determined by applying histogram equalization to all region histograms.
  • the computational complexity is quite high, but interpolation techniques are not needed. Using the same clipping level in the histogram calculated for each pixel may result in excessive noise enhancement as there will be extra enhancement in areas of the image where edge information is not available.
  • Patent application No. IN2986CHE2013A was found in the relevant searching.
  • the application relates to automatic CLAHE method and system used for image enhancement.
  • the clipping level is calculated automatically.
  • the processing steps of the method according to the invention are as follows: dividing the image into subimages with the NxN matrix; calculating the histogram of each subimage and the peak value of the histogram; calculating the nominal clipping level with a limit from 0 to peak value using the binary search method; clipping the histogram to the nominal clipping level in case the peak of the histogram is higher than the nominal clipping level; summing the number of pixels causing the overshoot; evenly redistribution of the collected pixels over the entire histogram; applying histogram equalization to redistributed subimage histograms; gray level mapping for each pixel in the input image and applying the output mapping to each of the pixels in said input image to obtain the enhanced image.
  • Patent application No. KR101684990B1 was found in the relevant searching.
  • Application relates to a method that enhances the image contrast via CLAHE.
  • the relevant method is used especially in the underwater target detection and includes the following processing steps: detecting the edges of a gray image originally corresponding to a color image with a four-way Sobel edge detector; conversion of color image from RGB domain to HSI domain with non-linear conversion; applying contrast enhancement on a luminance vector via the CLAHE algorithm; converting the optimized HSI back to RGB and applying constrained multiplication and converting the display range of R, G, B components from [0, 1] to [0, 255].
  • the Sobel filter is used to detect the edges, however, the histogram value of the pixels within the detected edges cannot be calculated separately for each pixel.
  • the object of the invention is to solve the above-mentioned disadvantages by being inspired by the current conditions.
  • the main purpose of the invention is to calculate the clipping level parameter specific to each pixel of the image based on the surrounding edge information in the real-time sliding window CLAHE algorithm.
  • Another purpose of the invention is to increase the contrast only in the areas where the detail (edge information) is present and to prevent the increase in the contrast In the undetailed areas of the image.
  • Another purpose of the invention is to prevent noise amplification by reducing the clipping level in areas where there is no detail in the image.
  • Figure 1 graphically shows the result of the histogram for the 129x129 sample window in the sliding window CLAHE algorithm according to the invention.
  • Figure 2 graphically shows the histogram obtained by clipping the histogram shown in figure 1 from the calculated clipping level and redistributing the clipped values to the histogram.
  • Figure 3 graphically shows the cumulative histogram obtained from the histogram shown in figure 2.
  • Figure 4 graphically shows the transfer function, map showing what the output value of the relevant pixel will be, obtained after applying histogram equalization to the cumulative histogram shown in figure 3.
  • Sliding window CLAHE is a method that provides contrast enhancement in the image by replacing the value of the image within the NxN size window around that pixel with the value calculated over the histogram for each pixel in the middle of a two-dimensional filter.
  • N should be an odd number. Numbers greater than 33 and 33 are preferred in order to obtain significant results. In the exemplary application of the invention, 129 is preferred.
  • Histogram is a function that shows the distribution of pixel values found in an image. For example, if there are 10 pixels with 0 value in a window, the histogram shows the value 10 for the value 0. Pixels can take integer values between 0 and 255 in an 8-bit image, therefore, a histogram has 256 values. Histogram is generally graphically shown.
  • the histogram of the pixels in the surrounding window is calculated for each pixel.
  • Calculation process is provided by adding the histogram of the column entering the sliding window and subtracting the histogram of the column exiting the sliding window, for each pixel consecutively.
  • Figure 1 the result of the histogram calculation for the 129x129 sample window is graphically shown.
  • the clipping level is calculated for each window separately. The following steps are used to calculate clipping levels:
  • edge threshold In order to provide real-time contrast improvement, three parameters are defined by the user: edge threshold, minimum clip- and clip alpha Maximum clipping value is obtained by adding the clip alpha to minimum clip.
  • the image is preferably passed through a 9x9 sobel filter, thus obtaining high- frequency parts of the image, namely edge information.
  • the sobel filter is the most common method used to obtain high frequency and edge information in image processing. Edge information can also be obtained with a different process.
  • a comparator compares the user-specified edge threshold with the high frequency obtained from the sobel filter.
  • the high-frequency image obtained with the sobel filter as a result of the comparison is converted into a binary image consisting of 0 and 1.
  • the conversion process facilitates the hardware implementation of the method according to the invention.
  • the total number of 1 ’s in the NxN window around each pixel in the resulting binary image is calculated. This calculation is performed with a method similar to histogram calculation. First, the pixels in the columns entering and exiting the window are recorded as inputs and outputs. Then, the column totals (inputs) entering the sliding window are added to the current total, and the totals (outgoing) of the columns leaving the sliding window are subtracted from the current total, and the total number of 1 ’s in the window is obtained therewith.
  • the parts that exceed the clipping level in the histogram are cut up to the clipping level, and the clipped pixels are equally redistributed to each element of the histogram.
  • the cumulative distribution of the clipped and redistributed histogram is calculated after the clipping process.
  • Figure 3 the cumulative histogram of the redistributed histogram is graphically shown.
  • l'[x][y] represents output pixel value
  • l[x][y] represents input pixel value
  • x and y represents pixel coordinates
  • h c iip P ed[i] represents clipped histogram
  • M represents total number of pixels in the window
  • i represents input pixel value and lower values
  • Nb represents the maximum element number a pixel can have.
  • Each pixel with 8-bit resolution has values in the range of 0 to 255, therefore, Nb is 256 at maximum.
  • a pixel-specific clipping level parameter is calculated according to edge information around each pixel for the real-time sliding window CLAHE algorithm.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Facsimile Image Signal Circuits (AREA)
  • Picture Signal Circuits (AREA)
  • Image Processing (AREA)

Abstract

The invention relates to a method for calculating the pixel-specific clipping level parameter based on the edge information around each pixel for the real-time sliding window CLAHE algorithm.

Description

VARIABLE CLIPPING LEVEL CALCULATION METHOD FOR CLAHE ALGORITHM
Technical field
The invention relates to a pixel-specific clipping level parameter calculation method for the real-time sliding window CLAHE (contrast-limited adaptive histogram equalization algorithm based on edge information around each pixel.
State of the Art
Today, real-time sliding window CLAHE algorithm is used to increase the contrast of the image in infrared or daytime imaging. In CLAHE algorithm, the amount of contrast increase is determined by a parameter called the clipping level. In the traditional sliding window CLAHE algorithm, the clipping level is applied to each pixel of the image with the same value, and it is not possible to reduce the clipping level separately in different areas without edge information. For this reason, it is not possible to avoid noise amplification caused by unnecessary contrast increase in areas without edge information such as the sky.
In the current method, a separate region histogram is calculated for all pixels in the image. Region histograms calculated for each pixel are clipped from a fixed clipping level, and values clipped from the region histogram are redistributed to the entire region histogram. Then, the output value of the relevant pixel is determined by applying histogram equalization to all region histograms. In this technique, the computational complexity is quite high, but interpolation techniques are not needed. Using the same clipping level in the histogram calculated for each pixel may result in excessive noise enhancement as there will be extra enhancement in areas of the image where edge information is not available.
Patent application No. IN2986CHE2013A was found in the relevant searching. The application relates to automatic CLAHE method and system used for image enhancement. In the automatic CLAHE method mentioned in the said application, the clipping level is calculated automatically. The processing steps of the method according to the invention are as follows: dividing the image into subimages with the NxN matrix; calculating the histogram of each subimage and the peak value of the histogram; calculating the nominal clipping level with a limit from 0 to peak value using the binary search method; clipping the histogram to the nominal clipping level in case the peak of the histogram is higher than the nominal clipping level; summing the number of pixels causing the overshoot; evenly redistribution of the collected pixels over the entire histogram; applying histogram equalization to redistributed subimage histograms; gray level mapping for each pixel in the input image and applying the output mapping to each of the pixels in said input image to obtain the enhanced image. In the relevant application, it is mentioned that the clipping level of the image is automatically calculated, however, there is no mention of calculating a separate clipping level for each pixel and using a sobel filter to achieve this. In the relevant application, "sliding window CLAHE” method is also not used.
Patent application No. KR101684990B1 was found in the relevant searching. Application relates to a method that enhances the image contrast via CLAHE. The relevant method is used especially in the underwater target detection and includes the following processing steps: detecting the edges of a gray image originally corresponding to a color image with a four-way Sobel edge detector; conversion of color image from RGB domain to HSI domain with non-linear conversion; applying contrast enhancement on a luminance vector via the CLAHE algorithm; converting the optimized HSI back to RGB and applying constrained multiplication and converting the display range of R, G, B components from [0, 1] to [0, 255]. In the application, it is mentioned that the Sobel filter is used to detect the edges, however, the histogram value of the pixels within the detected edges cannot be calculated separately for each pixel.
In conclusion, due to the problems mentioned above and the inadequacy of the existing solutions, it was necessary to make improvements in the relevant technical field.
Objects of the invention
The object of the invention is to solve the above-mentioned disadvantages by being inspired by the current conditions.
The main purpose of the invention is to calculate the clipping level parameter specific to each pixel of the image based on the surrounding edge information in the real-time sliding window CLAHE algorithm.
Another purpose of the invention is to increase the contrast only in the areas where the detail (edge information) is present and to prevent the increase in the contrast In the undetailed areas of the image.
Another purpose of the invention is to prevent noise amplification by reducing the clipping level in areas where there is no detail in the image. The structural and characteristic features of the present invention will be understood clearly by the following drawings and the detailed description made with reference to these drawings and therefore the evaluation shall be made by taking these figures and the detailed description into consideration.
Description of Figures
Figure 1 graphically shows the result of the histogram for the 129x129 sample window in the sliding window CLAHE algorithm according to the invention.
Figure 2 graphically shows the histogram obtained by clipping the histogram shown in figure 1 from the calculated clipping level and redistributing the clipped values to the histogram.
Figure 3 graphically shows the cumulative histogram obtained from the histogram shown in figure 2.
Figure 4 graphically shows the transfer function, map showing what the output value of the relevant pixel will be, obtained after applying histogram equalization to the cumulative histogram shown in figure 3.
Detailed Description of the Invention
This detailed description elaborates the preferred embodiments of the variable clipping level calculation method for the sliding window CLAHE algorithm according to the invention, for only a better understanding of the relevant subject.
Sliding window CLAHE is a method that provides contrast enhancement in the image by replacing the value of the image within the NxN size window around that pixel with the value calculated over the histogram for each pixel in the middle of a two-dimensional filter. N should be an odd number. Numbers greater than 33 and 33 are preferred in order to obtain significant results. In the exemplary application of the invention, 129 is preferred.
Histogram is a function that shows the distribution of pixel values found in an image. For example, if there are 10 pixels with 0 value in a window, the histogram shows the value 10 for the value 0. Pixels can take integer values between 0 and 255 in an 8-bit image, therefore, a histogram has 256 values. Histogram is generally graphically shown.
In the method according to the invention, firstly, the histogram of the pixels in the surrounding window is calculated for each pixel. Calculation process is provided by adding the histogram of the column entering the sliding window and subtracting the histogram of the column exiting the sliding window, for each pixel consecutively. In Figure 1 , the result of the histogram calculation for the 129x129 sample window is graphically shown. After calculating the histogram, the clipping level is calculated for each window separately. The following steps are used to calculate clipping levels:
In order to provide real-time contrast improvement, three parameters are defined by the user: edge threshold, minimum clip- and clip alpha
Figure imgf000005_0001
Maximum clipping value is obtained by adding the clip alpha to minimum clip.
The image is preferably passed through a 9x9 sobel filter, thus obtaining high- frequency parts of the image, namely edge information. The sobel filter is the most common method used to obtain high frequency and edge information in image processing. Edge information can also be obtained with a different process.
A comparator compares the user-specified edge threshold with the high frequency obtained from the sobel filter. The high-frequency image obtained with the sobel filter as a result of the comparison is converted into a binary image consisting of 0 and 1. The conversion process facilitates the hardware implementation of the method according to the invention.
The total number of 1 ’s in the NxN window around each pixel in the resulting binary image is calculated. This calculation is performed with a method similar to histogram calculation. First, the pixels in the columns entering and exiting the window are recorded as inputs and outputs. Then, the column totals (inputs) entering the sliding window are added to the current total, and the totals (outgoing) of the columns leaving the sliding window are subtracted from the current total, and the total number of 1 ’s in the window is obtained therewith.
In the next process step, the total number of 1 ’s obtained is proportioned to the number of pixels in this window. The value obtained as a result of this proportioning sets forth the edge density (X) around that pixel. This process is elaborated with the following equation:
Figure imgf000005_0002
After calculating the edge density, the clipping level value is calculated by adding the minimum clipping value to the value obtained by expanding the clip alpha value determined by the user with the edge density. This process is elaborated with the following equation: clipping level = a ^y + u
After the clipping level is determined by the abovementioned process steps, the parts that exceed the clipping level in the histogram are cut up to the clipping level, and the clipped pixels are equally redistributed to each element of the histogram.
The cumulative distribution of the clipped and redistributed histogram is calculated after the clipping process. In Figure 3, the cumulative histogram of the redistributed histogram is graphically shown.
Finally, the ratio of the cumulative histogram to the number of pixels in the window is obtained and this ratio is expanded with the maximum pixel value (255 for 8-bit image). This newly obtained value represents the transfer function. In this transfer function, the new pixel value corresponding to the processed pixel value is obtained. Obtaining new pixel values for each pixel is elaborated by the following equation:
Figure imgf000006_0001
In this formula, l'[x][y] represents output pixel value, l[x][y] represents input pixel value, x and y represents pixel coordinates, hciipPed[i] represents clipped histogram, M represents total number of pixels in the window, i represents input pixel value and lower values, and Nb represents the maximum element number a pixel can have. Each pixel with 8-bit resolution has values in the range of 0 to 255, therefore, Nb is 256 at maximum.
Due to the method according to the invention, a pixel-specific clipping level parameter is calculated according to edge information around each pixel for the real-time sliding window CLAHE algorithm.
In high-frequency areas with edge information in the image, increasing the clipping level in the regions within the window increases contrast and improves the viewing. In low- frequency areas where there is no edge information in the image, the amount of edge remaining in the window is low, so a low clipping level is applied so that the contrast in those areas does not reveal noise.

Claims

6
CLAIMS A method increasing contrast for the real time sliding window CLAHE algorithm, characterized by comprising the following process steps;
• calculating of the histogram of pixels in the surrounding window for each pixel, o determining the edge threshold, minimum cli and clip alpha parameters by the user, o obtaining high-frequency parts of the image via a filter, o comparing the edge threshold and the high frequency information obtained from the filter by a comparator, and consequently converting the image into a binary image, o
■ collection of all the pixels in the window
■ adding the sum of each column entering the sliding window to the current total, and
■ subtracting the sum of each column out of the sliding window from the current total, obtaining the total number of 1 ’s in the NxN window around each pixel in the resulting binary image via process steps, o calculating the edge density by obtaining the ratio of the total number of 1 to the number of pixels in the window, o adding the minimum clip value to the value obtained by expanding the clip alpha value with the edge density, calculation of the clipping level value for each pixel separately with the processing steps,
• parts of the histogram that exceed the clipping level are clipped from the clipping level, and pixels above the clipping level are evenly redistributed to each element of the histogram,
• calculation of cumulative histogram over the clipped and redistributed histogram, 7
• obtaining the transfer function by expanding the ratio of the cumulative histogram to the total number of pixels in the window by the maximum pixel value, and
• obtaining the output pixel value corresponding to the input pixel value via the transfer function. The method according to Claim 1 , characterized in that; sobel filter is used for obtaining the high-frequency parts of the image.
PCT/TR2021/050869 2020-08-28 2021-08-27 Variable clipping level calculation method for clahe algorithm WO2022046016A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP21862237.1A EP4091130A4 (en) 2020-08-28 2021-08-27 Variable clipping level calculation method for clahe algorithm

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
TR2020/13601 2020-08-28
TR2020/13601A TR202013601A1 (en) 2020-08-28 2020-08-28 Variable clipping level calculation method for the CLAHE algorithm

Publications (1)

Publication Number Publication Date
WO2022046016A1 true WO2022046016A1 (en) 2022-03-03

Family

ID=80353763

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/TR2021/050869 WO2022046016A1 (en) 2020-08-28 2021-08-27 Variable clipping level calculation method for clahe algorithm

Country Status (3)

Country Link
EP (1) EP4091130A4 (en)
TR (1) TR202013601A1 (en)
WO (1) WO2022046016A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116309572A (en) * 2023-05-19 2023-06-23 无锡康贝电子设备有限公司 Intelligent recognition method for numerical control machine tool components based on images

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116778261B (en) * 2023-08-21 2023-11-14 山东恒信科技发展有限公司 Raw oil grade classification method based on image processing

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120263366A1 (en) * 2011-04-14 2012-10-18 Zhimin Huo Enhanced visualization for medical images
US20160189354A1 (en) * 2014-12-26 2016-06-30 Ricoh Company, Ltd. Image processing system, image processing device, and image processing method
US20160335751A1 (en) * 2015-05-17 2016-11-17 Endochoice, Inc. Endoscopic Image Enhancement Using Contrast Limited Adaptive Histogram Equalization (CLAHE) Implemented In A Processor
KR101684990B1 (en) 2015-08-04 2016-12-12 청주대학교 산학협력단 Method for deblurring vehicle image using sigma variation of Bilateral Filter
CN107481202A (en) 2017-08-14 2017-12-15 深圳市华星光电半导体显示技术有限公司 A kind of method of dynamic range of images enhancing

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6613697B2 (en) * 2015-08-06 2019-12-04 株式会社リコー Image processing apparatus, program, and recording medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120263366A1 (en) * 2011-04-14 2012-10-18 Zhimin Huo Enhanced visualization for medical images
US20160189354A1 (en) * 2014-12-26 2016-06-30 Ricoh Company, Ltd. Image processing system, image processing device, and image processing method
US20160335751A1 (en) * 2015-05-17 2016-11-17 Endochoice, Inc. Endoscopic Image Enhancement Using Contrast Limited Adaptive Histogram Equalization (CLAHE) Implemented In A Processor
KR101684990B1 (en) 2015-08-04 2016-12-12 청주대학교 산학협력단 Method for deblurring vehicle image using sigma variation of Bilateral Filter
CN107481202A (en) 2017-08-14 2017-12-15 深圳市华星光电半导体显示技术有限公司 A kind of method of dynamic range of images enhancing

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4091130A4

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116309572A (en) * 2023-05-19 2023-06-23 无锡康贝电子设备有限公司 Intelligent recognition method for numerical control machine tool components based on images
CN116309572B (en) * 2023-05-19 2023-07-21 无锡康贝电子设备有限公司 Intelligent recognition method for numerical control machine tool components based on images

Also Published As

Publication number Publication date
EP4091130A4 (en) 2024-02-21
TR202013601A1 (en) 2022-03-21
EP4091130A1 (en) 2022-11-23

Similar Documents

Publication Publication Date Title
Kim et al. Optimized contrast enhancement for real-time image and video dehazing
CN103369209B (en) Vedio noise reduction device and method
US8279345B2 (en) System and method for random noise estimation in a sequence of images
JP4460839B2 (en) Digital image sharpening device
EP4091130A1 (en) Variable clipping level calculation method for clahe algorithm
US7570831B2 (en) System and method for estimating image noise
CN111210393B (en) Color image enhancement method, system and storage medium based on edge extraction
JP2003198850A (en) Digital color image processing method
CN107292834B (en) Infrared image detail enhancement method
CN103679173A (en) Method for detecting image salient region
Harb et al. Improved image magnification algorithm based on Otsu thresholding
Dharejo et al. A color enhancement scene estimation approach for single image haze removal
CN107833189A (en) The Underwater Target Detection image enchancing method of the limited self-adapting histogram equilibrium of contrast
CN113012061A (en) Noise reduction processing method and device and electronic equipment
KR100949403B1 (en) Image noise reduction method and apparatus
CN1329873C (en) Equalizing method for truncating histogram
CN110298796B (en) Low-illumination image enhancement method based on improved Retinex and logarithmic image processing
CN110111280A (en) A kind of enhancement algorithm for low-illumination image of multi-scale gradient domain guiding filtering
WO2020107308A1 (en) Low-light-level image rapid enhancement method and apparatus based on retinex
Messina et al. Image quality improvement by adaptive exposure correction techniques
CN103618904B (en) Motion estimation method and device based on pixels
Lai et al. Novel mean-shift based histogram equalization using textured regions
CN112085683B (en) Depth map credibility detection method in saliency detection
CN114140348A (en) Contrast enhancement method, device and equipment
Lavania et al. A comparative study of image enhancement using histogram approach

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21862237

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021862237

Country of ref document: EP

Effective date: 20220819

NENP Non-entry into the national phase

Ref country code: DE