[go: up one dir, main page]

CN104268845A - Self-adaptive double local reinforcement method of extreme-value temperature difference short wave infrared image - Google Patents

Self-adaptive double local reinforcement method of extreme-value temperature difference short wave infrared image Download PDF

Info

Publication number
CN104268845A
CN104268845A CN201410597724.9A CN201410597724A CN104268845A CN 104268845 A CN104268845 A CN 104268845A CN 201410597724 A CN201410597724 A CN 201410597724A CN 104268845 A CN104268845 A CN 104268845A
Authority
CN
China
Prior art keywords
image
histogram
self
local
temperature difference
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
CN201410597724.9A
Other languages
Chinese (zh)
Inventor
周津同
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
BEIJING CHINTONLIA TECHNOLOGY Co Ltd
Original Assignee
BEIJING CHINTONLIA TECHNOLOGY Co Ltd
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 BEIJING CHINTONLIA TECHNOLOGY Co Ltd filed Critical BEIJING CHINTONLIA TECHNOLOGY Co Ltd
Priority to CN201410597724.9A priority Critical patent/CN104268845A/en
Publication of CN104268845A publication Critical patent/CN104268845A/en
Pending legal-status Critical Current

Links

Landscapes

  • Image Processing (AREA)

Abstract

The invention discloses a self-adaptive double local reinforcement method of an extreme-value temperature difference short wave infrared image. The method comprises the following steps: step 1, cutting the image into a plurality of local images; step 2, performing K-means clustering cutting on each local image cut in the step 1, dividing the local image into a plurality layers of areas, taking the maximum value and the minimum value of the grey level of each area as the segmentation threshold values of multiple histograms; step 3, dividing the image into a plurality of grey areas according to the dividing threshold values in the step 2, performing histogram statistics on each grey area, and performing self-adaptive platform statistics on the segmented image histogram; step 4: outputting all the histogram combinations in a balanced manner to obtain a reinforced image. The self-adaptive double local reinforcement method is used for improving the quality of the extreme-value temperature infrared image and the local contrast ratio and strengthening the visual effect.

Description

The two local enhancement methods of self-adaptation of extreme value temperature difference short-wave infrared image
Technical field
The present invention relates to the technical field of infrared imaging, particularly relate to the two local enhancement methods of self-adaptation of extreme value temperature difference short-wave infrared image.
Background technology
The defects such as extreme value temperature difference short-wave infrared image is lost because the extreme value temperature difference of target presents local detail usually, and resolution is low, visual effect is fuzzy and signal to noise ratio (S/N ratio) is low.Usually use histogram equalization algorithm in extreme value temperature difference short-wave infrared field of image enhancement, gray-scale values many for pixel distribution is expanded to more tonal ranges by it, by gray compression few for pixel distribution to less tonal range.But the features such as resolution is low, contrast is low, image blur that infrared image has.Traditional histogram equalization algorithm has inborn defect in infrared image enhancement application.
For improving the defect of traditional histogram equalization algorithm, researchist proposes plateau equalization (PHE) on this basis, this algorithm adds the threshold value of statistical pixel number on the basis of histogram equalization, overcomes histogram equalization algorithm and strengthens excessive phenomenon by arranging enhancing degree to some intensity profile; Consider that the details of infrared image strengthens, researchist proposes local histogram equalization (LAHE), and many histogram equalizations (multi-histogram equalization, MHE) etc. that the people such as Menotti proposes.
Image is divided into multiple topography by LAHE algorithm, then carries out histogram equalization enhancing respectively; Enhance the local message of image to a certain extent.But for extreme value temperature-difference target, because essence adopts histogram equalization, details still can be lost.And histogram is resolved into multiple subgraph (sub-image) by MHE algorithm, carry out histogram equalization respectively, its algorithm calculates more complicated, and real-time implementation is more difficult.
Therefore, improve extreme value temperature infrared picture quality, improve local contrast, strengthen visual effect and be conducive to the application such as follow-up detection, identification and tracking, become a urgent problem.
Summary of the invention
The object of the present invention is to provide the two local enhancement methods of the self-adaptation of extreme value temperature difference short-wave infrared image to overcome the above-mentioned problems in the prior art.
For achieving the above object, the invention provides the two local enhancement methods of self-adaptation of extreme value temperature difference short-wave infrared image, comprising:
Step 1: image is cut into multiple topography;
Step 2: carry out the segmentation of K mean cluster to described each topography of segmentation in step 1, described topography is divided into multi-layer area, using the maximal value of the grey level in each region and minimum value as how histogrammic segmentation threshold;
Step 3: Iamge Segmentation is become multiple gray areas according to segmentation threshold described in step 2, and statistics with histogram is carried out to each gray areas, and adaptive platform statistics is carried out to the image histogram after segmentation;
Step 4: exported by all histogram combined equalizations, be enhanced image.
Preferably, clustering method is adopted image to be cut into 16 topographies in described step 1.
Preferably, described step 3 introduces histogram local distribution information on the histogrammic basis of adaptive platform, can retain the mean flow rate of each Gray Level Segments of image on the basis strengthening image detail.Avoid the defects such as the gray scale occurred in histogram enhancement excessively strengthens, visual effect is unnatural.
Preferably, in described step 4,16 width topographies are strengthened respectively, and occur 16 different enhancing results, adopt bilinear interpolation algorithm to do last gradation of image and splice;
Suppose f 00, f 01, f 10, f 11be corresponding histogram look-up table in adjacent four the topography's blocks of pixel, (x i, y jrepresent interpolation point coordinate in X-axis and Y-axis, in i=1 ~ N-1, j=1 ~ N-1 design, N gets 2 k, k=1,2,3...), bilinear interpolation algorithm be exactly according to the gray-scale value of these four points calculate insert in the middle of certain any gray-scale value
Algorithmic formula is as follows:
f x i 0 = f 00 + x i [ f 10 - f 00 ]
f x i 1 = f 01 + x i [ f 11 - f 01 ] f x i y i = f x i 0 + y j [ f x i 1 - f x i 0 ]
= x i [ f 10 - f 00 ] + y j [ f 01 - f 00 ] + x i y j [ f 11 + f 00 - f 01 - f 10 ] + f 00 .
The present invention is in conjunction with plateau equalization (PHE), the advantage of local histogram equalization (LAHE) and many histogram equalizations (MHE), and introduce space and statistical information, the method of the two local enhancement of a kind of self-adaptation for extreme value temperature difference image is proposed, the method is from the target distribution of infrared image, introduce image space information, from analysis local space gray-scale watermark, image region segmentation is incorporated among algorithm, the mean flow rate of image can be retained on the basis strengthening image detail, improve efficiently the deficiency of said method, give full play to the advantage of said method, various details in extreme value temperature difference short-wave infrared image are presented best visual effect.
Accompanying drawing illustrates:
Fig. 1 is the schematic flow sheet of the two local enhancement methods of short-wave infrared image.
Embodiment:
For making object of the invention process, technical scheme and advantage clearly, below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is further described in more detail.In the accompanying drawings, same or similar label represents same or similar element or has element that is identical or similar functions from start to finish.Described embodiment is the present invention's part embodiment, instead of whole embodiments.Be exemplary below by the embodiment be described with reference to the drawings, be intended to for explaining the present invention, and can not limitation of the present invention be interpreted as.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.Below in conjunction with accompanying drawing, embodiments of the invention are described in detail.
In describing the invention; it will be appreciated that; term " " center ", " longitudinal direction ", " transverse direction ", "front", "rear", "left", "right", " vertically ", " level ", " top ", " end " " interior ", " outward " etc. instruction orientation or position relationship be based on orientation shown in the drawings or position relationship; be only the present invention for convenience of description and simplified characterization; instead of instruction or imply indication device or element must have specific orientation, with specific azimuth configuration and operation, therefore can not be interpreted as limiting the scope of the invention.
According to the two local enhancement methods of the self-adaptation of the extreme value temperature difference short-wave infrared image of the present invention one broad embodiment, comprising:
Step 1: image is cut into multiple topography;
Step 2: carry out the segmentation of K mean cluster to described each topography of segmentation in step 1, described topography is divided into multi-layer area, using the maximal value of the grey level in each region and minimum value as how histogrammic segmentation threshold;
Step 3: Iamge Segmentation is become multiple gray areas according to segmentation threshold described in step 2, and statistics with histogram is carried out to each gray areas, and adaptive platform statistics is carried out to the image histogram after segmentation;
Step 4: exported by all histogram combined equalizations, be enhanced image.
For scene internal object and the larger infrared image of the background temperature difference, owing to having the large feature of target and background imaging gray difference, the infrared image histogram enhancement algorithm of the direct overall situation strengthens entire image, have ignored image information between target background, target and background in image is excessively strengthened, and the target under other backgrounds may be fogged by annihilation, be difficult to obtain desirable effect.Therefore, for the infrared image of the extreme value temperature difference, be first that it is cut into several topographies, strengthen respectively after target and background is subdivided into multiple topography.Be enhanced same histogram from spatially avoiding extreme value temperature difference image like this.
As shown in Figure 1, processing power according to FPGA is divided into 16 topographies the image that resolution is 384*288, the size of each topography is 96*72 pixel, after being divided into topography, extreme value temperature difference image is segmented in one or more region, the gray difference of each like this topography can obviously reduce, if the object using traditional histogram enhancement algorithm can play each topography's details to highlight, but due to the out-of-shape of extreme value temperature-difference target or background, so simply to the shape being divided into topography and can not segmenting very well extreme value target and background, need better segmentation object in spatial character.
In order to segment the target and background of the extreme value temperature difference in topography, clustering method is adopted to carry out splitting again of topography in step 1.Clustering method is a kind of without supervised recognition learning method, from initial cluster center according to similarity and adjacency structural classification device.Thus given data object is divided into the class of some different s.Data set after cluster has the advantages that in class, between object height correlativity and class, object difference is larger, thus may be used for tagsort, also may be used for the segmentation of image.
Described step 3 introduces histogram local distribution information on the histogrammic basis of adaptive platform, can retain the mean flow rate of each Gray Level Segments of image on the basis strengthening image detail.Avoid the defects such as the gray scale occurred in histogram enhancement excessively strengthens, visual effect is unnatural.
In described step 4,16 width topographies are strengthened respectively, and occur 16 different enhancing results, adopt bilinear interpolation algorithm to do last gradation of image and splice;
Suppose f 00, f 01, f 10, f 11be corresponding histogram look-up table in adjacent four the topography's blocks of pixel, (x i, y jrepresent interpolation point coordinate in X-axis and Y-axis, in i=1 ~ N-1, j=1 ~ N-1 design, N gets 2 k, k=1,2,3...), bilinear interpolation algorithm be exactly according to the gray-scale value of these four points calculate insert in the middle of certain any gray-scale value
Algorithmic formula is as follows:
f x i 0 = f 00 + x i [ f 10 - f 00 ]
f x i 1 = f 01 + x i [ f 11 - f 01 ] f x i y i = f x i 0 + y j [ f x i 1 - f x i 0 ]
= x i [ f 10 - f 00 ] + y j [ f 01 - f 00 ] + x i y j [ f 11 + f 00 - f 01 - f 10 ] + f 00 .
Finally it is to be noted: above embodiment only in order to technical scheme of the present invention to be described, is not intended to limit.Although with reference to previous embodiment to invention has been detailed description, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein portion of techniques feature; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.

Claims (4)

1. the two local enhancement methods of the self-adaptation of extreme value temperature difference short-wave infrared image, is characterized in that, comprising:
Step 1: image is cut into multiple topography;
Step 2: carry out the segmentation of K mean cluster to described each topography of segmentation in step 1, described topography is divided into multi-layer area, using the maximal value of the grey level in each region and minimum value as how histogrammic segmentation threshold;
Step 3: Iamge Segmentation is become multiple gray areas according to segmentation threshold described in step 2, and statistics with histogram is carried out to each gray areas, and adaptive platform statistics is carried out to the image histogram after segmentation;
Step 4: exported by all histogram combined equalizations, be enhanced image.
2. the two local enhancement methods of the self-adaptation of extreme value temperature difference short-wave infrared image as claimed in claim 1, is characterized in that, adopt clustering method image to be cut into 16 topographies in described step 1.
3. the two local enhancement methods of the self-adaptation of extreme value temperature difference short-wave infrared image as claimed in claim 1, it is characterized in that, described step 3 introduces histogram local distribution information on the histogrammic basis of adaptive platform, the mean flow rate of each Gray Level Segments of image can be retained on the basis strengthening image detail, avoid the defects such as the gray scale occurred in histogram enhancement excessively strengthens, visual effect is unnatural.
4. the two local enhancement methods of the self-adaptation of extreme value temperature difference short-wave infrared image as claimed in claim 2, it is characterized in that, in described step 4,16 width topographies are strengthened respectively, and occur 16 different enhancing results, adopt bilinear interpolation algorithm to do last gradation of image and splice;
Suppose f 00, f 01, f 10, f 11be corresponding histogram look-up table in adjacent four the topography's blocks of pixel, (x i, y jrepresent interpolation point coordinate in X-axis and Y-axis, in i=1 ~ N-1, j=1 ~ N-1 design, N gets 2 k, k=1,2,3...), bilinear interpolation algorithm be exactly according to the gray-scale value of these four points calculate insert in the middle of certain any gray-scale value
Algorithmic formula is as follows:
f x i 0 = f 00 + x i [ f 10 - f 00 ] f x i 1 = f 01 + x i [ f 11 - f 01 ] f x i y i = f x i 0 + y i [ f x i 1 - f x i 0 ]
= x i [ f 10 - f 00 ] + y i [ f 01 - f 00 ] + x i y j [ f 11 + f 00 - f 01 - f 10 ] + f 00 .
CN201410597724.9A 2014-10-31 2014-10-31 Self-adaptive double local reinforcement method of extreme-value temperature difference short wave infrared image Pending CN104268845A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410597724.9A CN104268845A (en) 2014-10-31 2014-10-31 Self-adaptive double local reinforcement method of extreme-value temperature difference short wave infrared image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410597724.9A CN104268845A (en) 2014-10-31 2014-10-31 Self-adaptive double local reinforcement method of extreme-value temperature difference short wave infrared image

Publications (1)

Publication Number Publication Date
CN104268845A true CN104268845A (en) 2015-01-07

Family

ID=52160365

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410597724.9A Pending CN104268845A (en) 2014-10-31 2014-10-31 Self-adaptive double local reinforcement method of extreme-value temperature difference short wave infrared image

Country Status (1)

Country Link
CN (1) CN104268845A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105654438A (en) * 2015-12-27 2016-06-08 西南技术物理研究所 Gray scale image fitting enhancement method based on local histogram equalization
CN109584189A (en) * 2017-09-28 2019-04-05 中国科学院长春光学精密机械与物理研究所 The real time enhancing method and device of soft image
CN109813757A (en) * 2019-02-18 2019-05-28 中国石油大学(北京) Method and device for extracting fault infrared thermal imaging features of shale gas fracturing equipment
CN111261088A (en) * 2020-02-25 2020-06-09 京东方科技集团股份有限公司 Image drawing method and device and display device
CN114549670A (en) * 2022-02-23 2022-05-27 京东方数字科技有限公司 Image processing method and image processing system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102208101A (en) * 2011-04-29 2011-10-05 中国航空工业集团公司洛阳电光设备研究所 Self-adaptive linearity transformation enhancing method of infrared image
US20120263366A1 (en) * 2011-04-14 2012-10-18 Zhimin Huo Enhanced visualization for medical images

Patent Citations (2)

* 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
CN102208101A (en) * 2011-04-29 2011-10-05 中国航空工业集团公司洛阳电光设备研究所 Self-adaptive linearity transformation enhancing method of infrared image

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
贺明等: "极值温差红外图像的自适应双局部增强算法", 《强激光与粒子束》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105654438A (en) * 2015-12-27 2016-06-08 西南技术物理研究所 Gray scale image fitting enhancement method based on local histogram equalization
CN109584189A (en) * 2017-09-28 2019-04-05 中国科学院长春光学精密机械与物理研究所 The real time enhancing method and device of soft image
CN109813757A (en) * 2019-02-18 2019-05-28 中国石油大学(北京) Method and device for extracting fault infrared thermal imaging features of shale gas fracturing equipment
CN109813757B (en) * 2019-02-18 2020-04-24 中国石油大学(北京) Shale gas fracturing equipment fault infrared thermal imaging feature extraction method and device
CN111261088A (en) * 2020-02-25 2020-06-09 京东方科技集团股份有限公司 Image drawing method and device and display device
CN111261088B (en) * 2020-02-25 2023-12-12 京东方科技集团股份有限公司 Image drawing method and device and display device
CN114549670A (en) * 2022-02-23 2022-05-27 京东方数字科技有限公司 Image processing method and image processing system

Similar Documents

Publication Publication Date Title
CN109961049B (en) Cigarette brand identification method under complex scene
CN109918969B (en) Face detection method and device, computer device and computer readable storage medium
CN101783012B (en) An Automatic Image Dehazing Method Based on Dark Channel Color
CN107016677B (en) A cloud image segmentation method based on FCN and CNN
CN116681636B (en) Light infrared and visible light image fusion method based on convolutional neural network
CN108197606A (en) The recognition methods of abnormal cell in a kind of pathological section based on multiple dimensioned expansion convolution
CN110033471A (en) A kind of wire detection method based on connected domain analysis and morphological operation
CN112184585B (en) An image completion method and system based on semantic edge fusion
CN114332650B (en) Remote sensing image road identification method and system
CN109241973B (en) A fully automatic soft segmentation method of characters under texture background
CN102289668A (en) Binaryzation processing method of self-adaption word image based on pixel neighborhood feature
CN111583279A (en) A Superpixel Image Segmentation Method Based on PCBA
CN111986126B (en) Multi-target detection method based on improved VGG16 network
CN104616308A (en) Multiscale level set image segmenting method based on kernel fuzzy clustering
CN104268845A (en) Self-adaptive double local reinforcement method of extreme-value temperature difference short wave infrared image
CN105550999A (en) Video image enhancement processing method based on background reuse
CN118297837B (en) Infrared simulator virtual image enhancement system based on image processing
CN103020953A (en) Segmenting method of fingerprint image
CN111062331A (en) Mosaic detection method and device for image, electronic equipment and storage medium
CN112749599B (en) Image enhancement method, device and server
CN114240988B (en) Image segmentation method based on nonlinear scale space
Kuzovkin et al. Descriptor-based image colorization and regularization
CN111027564A (en) Low-illumination imaging license plate recognition method and device based on deep learning integration
CN103093241B (en) Based on the remote sensing image nonuniformity cloud layer method of discrimination of homogeneity process
CN111429436B (en) An essential image analysis method based on multi-scale attention and label loss

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20150107

RJ01 Rejection of invention patent application after publication