CN103440661A - Micrometer-wood-fiber diameter detection algorithm based on HSV space and area selection - Google Patents

Micrometer-wood-fiber diameter detection algorithm based on HSV space and area selection Download PDF

Info

Publication number
CN103440661A
CN103440661A CN2013103980454A CN201310398045A CN103440661A CN 103440661 A CN103440661 A CN 103440661A CN 2013103980454 A CN2013103980454 A CN 2013103980454A CN 201310398045 A CN201310398045 A CN 201310398045A CN 103440661 A CN103440661 A CN 103440661A
Authority
CN
China
Prior art keywords
diameter
algorithm
fiber
micro
wood fiber
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.)
Granted
Application number
CN2013103980454A
Other languages
Chinese (zh)
Other versions
CN103440661B (en
Inventor
齐红
任洪娥
刘冕
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.)
Northeast Forestry University
Original Assignee
Northeast Forestry University
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 Northeast Forestry University filed Critical Northeast Forestry University
Priority to CN201310398045.4A priority Critical patent/CN103440661B/en
Publication of CN103440661A publication Critical patent/CN103440661A/en
Application granted granted Critical
Publication of CN103440661B publication Critical patent/CN103440661B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Image Analysis (AREA)

Abstract

Disclosed is a micrometer-wood-fiber diameter detection algorithm based on HSV space and area selection. The algorithm includes the following steps: shooting a microscopic magnification image of micrometer wood fiber and converting an RGB image into an HSV color space; carrying out binaryzation based on an S component; selectively eliminating irregular-form characteristics such as branches and rags and the like based on user demands; extracting a main-body area of the wood fiber and testing an operator through use of morphological single-scale edges so that edges of the wood fiber are obtained; and obtaining a diameter of the wood fiber through use of an improved segmented-fiber measuring method based on a maximum-inscribed-circle algorithm on the image, the edges of which are extracted. The micrometer-wood-fiber diameter detection algorithm based on the HSV space and area selection not only makes full use of color information of the image but also detects the diameter based on the maximum-inscribed-circle algorithm so that not only a traditional idea of constructing a liner equation is abandoned but also rotation and translation of the wood-fiber image are overcome and effects of the fiber rags on diameter detection are reduced as far as possible. Therefore, the algorithm has great significance to guidance of simulation studies and realization of diameter intelligent detection of the micrometer wood fiber.

Description

Based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm
Affiliated technical field
The present invention relates to a kind of Micro-wood Fiber examine of diameter algorithm based on micro-image, relate in particular to a kind of based on hsv color space and regioselective improved Micro-wood Fiber examine of diameter algorithm.
Background technology
The fibre diameter detection method has pneumatic tester method, optical microscope sciagraphy, optical microscope autoscan method, laser scanning method, scanning electron microscopy etc. at present, but because of instrument expensive, range of application is limited, can not be generalized in production well, thereby the difficulty that the robotization of xylogen diameter detects is increased.RGB color model commonly used is owing to there being very strong correlativity between R, G, B three-component, and the hsv color space more meets the feature of human eye to Color perception, and its three components are separate, stronger robustness is arranged.The people such as Wang Xiali are in document 1 " a kind of vehicle license extracting method based on the hsv color space " (computer engineering, 2004,30 (17)) in, utilize the hsv color space directly to be processed the coloured image that contains license plate of input, design a kind of car plate method of inspection, realize the car plate location.Mathematical morphology has perfect Fundamentals of Mathematics, obtained good effect in the rim detection field, as Li Jie waits in document 2 " research of the edge detection algorithm based on mathematical morphology " (computer science, 2012,39 (6A)) in by a kind of improved morphological operator, effectively detect edge, and kept edge-smoothing.Fiber areal survey method can be utilized the feature of each segmentation well, Zhang Peng flies to wait the people in document 3 " a kind of method based on the pretreated quick test fibre diameter of image " (Shanghai Textile science and technology, 2011, (39) 12) in, the wool fiber length direction, every about 9 pixels, fiber is carried out to the diameter measurement based on fitting a straight line from fiber one end, this algorithm has shortened the examine of diameter time, improved measurement efficiency, the advantage of performance digital image processing techniques.The application of maximum inscribed circle is more, the people such as Wu Wei are at document 4 " selection and the Position Research of area-of-interest in palm vein identification " (photoelectron laser, 2013, (24) 1) in, the ROI localization method of proposition based on thenar area, near wrist side maximum inscribed circle, to determine two invariant feature points on palm, utilize these two invariant feature points to determine ROI, ROI image to thenar area and centre of the palm district carries out the recognition performance contrast experiment, result shows, this method make multiple recognition performance index all be improved significantly.The above is based on hsv color space and regioselective Micro-wood Fiber examine of diameter algorithm theoretical foundation is provided.
Summary of the invention
The purpose of this invention is to provide a kind of based on HSV space and regioselective improved Micro-wood Fiber examine of diameter algorithm, this algorithm adopts the S component realize target based on the hsv color space to extract, based on the user's request concept, calculate the diameter of each segmentation and obtain mean value by improve traditional areal survey method based on the maximum inscribed circle algorithm, realizing the Micro-wood Fiber examine of diameter.Algorithm is not subject to the noise effect of background, and examine of diameter is effective, and applicability is stronger.
In order to achieve the above object, the present invention adopts following technical scheme:
Based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm, comprise the following steps:
(1) read in the micro-enlarged drawing of Micro-wood Fiber, the RGB image conversion that obtains, to the hsv color space, is extracted to the S component and carries out Threshold segmentation, carry out the operations such as mathematical morphology and filtering and noise reduction, obtain the binaryzation target of removing background;
(2) if the Micro-wood Fiber form is irregular, first carry out the user area selection, extract the xylogen body region, and carry out morphology single scale rim detection;
(3) the improved segmentation fibre diameter that the image after the edge detection carries out based on the maximum inscribed circle algorithm is measured, and the net result using the mean value of each sectional area diameter as the Micro-wood Fiber diameter, and algorithm finishes.
The accompanying drawing explanation
Fig. 1 is based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm flow chart;
Fig. 2 is the micro-enlarged image of Micro-wood Fiber of experiment use;
Fig. 3 is xylogen image hsv color space three-component figure, and wherein figure (a) divides spirogram for S divides spirogram, figure (c) for V for H divides spirogram, figure (b);
Fig. 4 is based on the S component xylogen image is carried out to binaryzation extraction result;
Fig. 5 carries out the extraction of xylogen body region to binary image based on user's request;
Fig. 6 carries out the result of rim detection to the image after extracted region;
Fig. 7 carries out the result of one of them areal survey based on maximum inscribed circle to xylogen;
Embodiment:
Fig. 1 is based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm flow chart; Of the present inventionly based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm, comprise the following steps:
(1) by the RGB image conversion that collects to the hsv color space;
(2) extract the S component and carry out the extraction of binaryzation target;
(3) carry out the operations such as mathematical morphology and filtering and noise reduction, obtain the binaryzation target of removing background;
(4) extract the Micro-wood Fiber body region based on user's request;
(5) adopt morphology single scale operator to detect the Micro-wood Fiber edge;
(6) image after the edge extraction adopts the improved areal survey method detection fibers diameter based on the maximum inscribed circle algorithm;
(7) to each segmentation diameter calculating mean value, as the net result of Micro-wood Fiber diameter.
Fig. 2 is the micro-enlarged image of Micro-wood Fiber collected;
Fig. 3 is xylogen image hsv color space three-component figure, and wherein figure (a) divides spirogram for S divides spirogram, figure (c) for V for H divides spirogram, figure (b).Because the S component has highlighted complete xylogen particle preferably, therefore adopt the S component to remove the background extracting target;
Fig. 4 is for to carry out based on S spatial extraction target the micro-enlarged image of xylogen by above-mentioned analysis, the figure as a result that the operations such as the line number of going forward side by side morphology and filtering and noise reduction obtain.From this interpretation, the method can successfully be isolated preferably xylogen from background;
Fig. 5 be to gained binary segmentation result image based on the user's request concept, remove the figure as a result of the irregular morphological feature of xylogen (such as bifurcated, burr etc.).Can find out from experimental result, based on user's request, not only obtain the xylogen body region of comparatively complete rule, be convenient to ensuing diameter measurement, and introduce user's participation, make algorithm more there is interactivity;
Fig. 6 is that the image after above-mentioned target is extracted carries out morphology single scale operator edge detection, and the operator of employing is: G ( f ( x , y ) ) = ( f ( x , y ) ⊕ b ( i , j ) ) - ( f ( x , y ) Θb ( i , j ) ) , Wherein, f (x, y) is pixel, and b (i, j) is structural elements.Algorithm picks disc structure unit, radius is 3;
Fig. 7 carries out the result of one of them areal survey based on maximum inscribed circle to xylogen.

Claims (4)

1. based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm, comprise the following steps:
(1) read in the micro-enlarged drawing of Micro-wood Fiber, the RGB image conversion that obtains, to the hsv color space, is extracted to the S component and carries out Threshold segmentation, carry out the operations such as mathematical morphology and filtering and noise reduction, obtain the binaryzation target of removing background;
(2) if the Micro-wood Fiber form is irregular, first carry out the user area selection, extract the xylogen body region, and carry out morphology single scale rim detection;
(3) the improved segmentation fibre diameter that the image after the edge detection carries out based on the maximum inscribed circle algorithm is measured, and the net result using the mean value of each sectional area diameter as the Micro-wood Fiber diameter, and algorithm finishes.
2. according to claim 1ly based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm, it is characterized in that, the RGB image of described step (1) is according to formula transform to the hsv color space, choose the S component that the target-to-background contrast is larger and carry out binary conversion treatment.
3. according to claim 1ly based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm, it is characterized in that, the user-selected area of described step (2) is removed the xylogen burr, and morphology single scale edge detection operator is:
Figure FDA0000377445640000012
wherein, f (x, y) is pixel, and b (i, j) is structural elements.Algorithm adopts disc structure unit, and radius is 3.
4. according to claim 1 based on HSV space and regioselective Micro-wood Fiber examine of diameter algorithm, it is characterized in that, the improved segmentation fibre diameter mensuration based on maximum inscribed circle of described step (3), according to user's request by the fiber segmentation, and sectional area is equivalent to polygon, obtain the diameter of this minute section fibre by the diameter that calculates this polygon maximum inscribed circle, each segmentation diameter is averaged as the diameter of micrometer fibers.
CN201310398045.4A 2013-09-05 2013-09-05 Based on HSV space and regioselective Micro-wood Fiber diameter detection algorithm Expired - Fee Related CN103440661B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310398045.4A CN103440661B (en) 2013-09-05 2013-09-05 Based on HSV space and regioselective Micro-wood Fiber diameter detection algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310398045.4A CN103440661B (en) 2013-09-05 2013-09-05 Based on HSV space and regioselective Micro-wood Fiber diameter detection algorithm

Publications (2)

Publication Number Publication Date
CN103440661A true CN103440661A (en) 2013-12-11
CN103440661B CN103440661B (en) 2016-08-17

Family

ID=49694354

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310398045.4A Expired - Fee Related CN103440661B (en) 2013-09-05 2013-09-05 Based on HSV space and regioselective Micro-wood Fiber diameter detection algorithm

Country Status (1)

Country Link
CN (1) CN103440661B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110853000A (en) * 2019-10-30 2020-02-28 北京中交国通智能交通系统技术有限公司 Detection method of track
CN115597507A (en) * 2022-10-27 2023-01-13 中钢集团郑州金属制品研究院股份有限公司（Cn） Diameter detection method of synthetic fiber

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102496144A (en) * 2011-11-15 2012-06-13 哈尔滨工程大学 NSCT (nonsubsampled contourlet transform) sonar image enhancement method based on HSV (hue, saturation and value) color space
CN102496157A (en) * 2011-11-22 2012-06-13 上海电力学院 Image detection method based on Gaussian multi-scale transform and color complexity
CN102930547A (en) * 2012-11-13 2013-02-13 中国农业大学 Cotton foreign fiber image online segmentation method and system on the condition of wind power delivery
CN103268615A (en) * 2013-06-05 2013-08-28 东北林业大学 Wood powder edge detection algorithm based on target extraction and improved morphological operator

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102496144A (en) * 2011-11-15 2012-06-13 哈尔滨工程大学 NSCT (nonsubsampled contourlet transform) sonar image enhancement method based on HSV (hue, saturation and value) color space
CN102496157A (en) * 2011-11-22 2012-06-13 上海电力学院 Image detection method based on Gaussian multi-scale transform and color complexity
CN102930547A (en) * 2012-11-13 2013-02-13 中国农业大学 Cotton foreign fiber image online segmentation method and system on the condition of wind power delivery
CN103268615A (en) * 2013-06-05 2013-08-28 东北林业大学 Wood powder edge detection algorithm based on target extraction and improved morphological operator

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
任洪娥等: "基于改进形态学算子的木粉边缘检测算法", 《计算机工程与应用》 *
伊海燕: "基于图像分析的羊绒纤维特征参数提取技术的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑(月刊)》 *
齐英杰等: "微米木纤维图像的模拟再现理论和应用研究", 《林业科技》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110853000A (en) * 2019-10-30 2020-02-28 北京中交国通智能交通系统技术有限公司 Detection method of track
CN110853000B (en) * 2019-10-30 2023-08-11 北京中交国通智能交通系统技术有限公司 Rut detection method
CN115597507A (en) * 2022-10-27 2023-01-13 中钢集团郑州金属制品研究院股份有限公司（Cn） Diameter detection method of synthetic fiber

Also Published As

Publication number Publication date
CN103440661B (en) 2016-08-17

Similar Documents

Publication Publication Date Title
CN111256594B (en) Method for measuring physical characteristics of surface state of aircraft skin
CN103839268B (en) Method for detecting fissure on surface of subway tunnel
Arachchige et al. An image processing algorithm for accurate extraction of the centerline from human metaphase chromosomes
CN101806583B (en) Microscopic image-based fiber fineness measurement method
CN100592323C (en) Method for identifying fingerprint facing image quality
JP2014524070A5 (en)
CN104034733A (en) Service life prediction method based on binocular vision monitoring and surface crack image recognition
CN101510295B (en) Design method for machine vision system based on PCIe and Vision Assistan
CN111174961A (en) Modal analysis-based cable force optical measurement method and measurement system thereof
CN102305795A (en) Method for positioning tiny crack on surface of concrete
CN113096121B (en) Pavement crack detection method and system based on cross fracture mechanics and image processing
CN102289806B (en) Method for measuring image definition by utilizing multi-scale morphological characteristics
CN103793709A (en) Cell recognition method and device, and urine analyzer
CN105869168A (en) Multi-source remote sensing image shape registering method based on polynomial fitting
CN103440661A (en) Micrometer-wood-fiber diameter detection algorithm based on HSV space and area selection
Cao et al. Power line detection based on symmetric partial derivative distribution prior
Michelin et al. Embryo cell membranes reconstruction by tensor voting
Marcolin et al. Three-dimensional face analysis via new geometrical descriptors
Soni et al. A Robust Vein Pattern-based Recognition System.
CN102708363A (en) Self-adaptive fingerprint identification system
Zhou et al. 3d dendrite reconstruction and spine identification
Lv et al. Maximizing structural similarity in multimodal biomedical microscopic images for effective registration
Hildebrandt et al. An enhanced feature set for pattern recognition based contrast enhancement of contact-less captured latent fingerprints in digitized crime scene forensics
CN103268615A (en) Wood powder edge detection algorithm based on target extraction and improved morphological operator
CN202711268U (en) Self-adaptive fingerprint identification system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20160817

Termination date: 20170905