CN103177241A - Method for positioning spacers of transmission lines by aid of video image processing technology - Google Patents

Abstract

The invention discloses a method for identifying and positioning spacers of transmission lines by the aid of a video image processing technology, and belongs to the technical field of digital video image processing and online monitoring for transmission lines. The method includes positioning target areas of images and positioning centers of the spacers by processes for normalizing the original images, acquiring direction fields, matching templates and the like; computing the direction fields of the images by a square-gradient algorithm when the target areas are positioned, and combining the direction fields with isotropic consistency information to position the target areas of the images corresponding to the transmission lines; and positioning the centers of the spacers by a histogram-based and normalized-correlation quick template matching algorithm. The quick template matching algorithm sets histogram matching joint threshold values and normalized-correlation matching maximum values as the foundation for judging the matching optimal positions, coordinates with the maximum probabilities in a plurality of matching results are selected to be used as center points of the spacers, and reference information is provided for monitoring conditions of the transmission lines and predicting faults.

Description

Utilize video image processing technology to locate the method for transmission line of electricity conductor spacer

Technical field

The invention belongs to that digital video image is processed and transmission line of electricity on-line monitoring technique field.Be particularly related to a kind of method of utilizing video image processing technology to locate the transmission line of electricity conductor spacer, identification and the location of the digital remote video image when transmission line of electricity is carried out on-line monitoring specifically, comprise to normalization, the field of direction of transmission line of electricity conductor spacer image ask for, a series of images treatment technology such as template matches, can automatically process and identify transmission line of electricity conductor spacer image, extract the positional information of conductor spacer in transmission line of electricity.

Background technology

The transmission line of electricity conductor spacer is the fastening between split conductor, makes to keep certain spacing distance between the sub-conductor of split conductor.As one of visual plant of transmission line of electricity, conductor spacer be obtain conducting wire sag, wave and line under the reliable object of reference of the state parameter such as growth-gen.

At present transmission line of electricity conductor spacer positional information being extracted the technological means that adopts, is mainly to adopt transit or ground laser radar technology.And existing monitoring means needs deeply operation on the spot of staff, be difficult to accurately complete monitoring in the bumpy highway section that the operating personnel is difficult to arrive, and the test duration is long.Adopting image processing techniques to identify the location to the transmission line of electricity conductor spacer is a kind of very directly perceived, easy method, and Chinese scholars has begun correlative study.Wherein, the transmission line of electricity image of the people such as Wai Ho Li take ordinary camera from transmission line of electricity side shooting is as goal in research, transmission pressure is considered as straight line and adopts template matching method that it is positioned, the characteristics of contact are connected between conductor spacer and wire in combining image, adopt Gabor filtering to realize the conductor spacer location, this algorithm can access desirable positioning result for low coverage, the simple image of background, but at the transmission line of electricity sag poor effect during large and background complexity.The video image that the people such as Li Junfang propose to be arranged on the video camera shooting on shaft tower is research object, adopt the algorithm of normalization relevant matches, carry out the coupling location of conductor spacer in conjunction with the pyramid hierarchical search, the method adopts both full-pixel normalization coupling to calculate, the algorithm calculated amount is large, consuming time higher, causes the matching algorithm poor in timeliness.

The present invention is based on digital remote video image technology, the video camera that employing is arranged on shaft tower is taken the transmission line of electricity live video, the digital picture that intercepts in the transmission line of electricity video flowing that is sent to Surveillance center is as research object, adopt a kind of new algorithm to realize automatic identification and location to the transmission line of electricity conductor spacer, the method is located the transmission line of electricity target area by the field of direction, and histogram and Normalized Cross Correlation Algorithm are combined, be used for the conductor spacer centralized positioning.This algorithm can be located the conductor spacer in transmission line of electricity more in real time, accurately, provides safely a kind of means of more precise and high efficiency for guaranteeing electrical production, and very important realistic meaning is arranged.

Summary of the invention

The purpose of this invention is to provide a kind of method of utilizing video image processing technology to locate the transmission line of electricity conductor spacer, it is characterized in that, the method is a kind of identification of transmission line of electricity conductor spacer and localization method based on the digital remote video image processing technology; By being erected at camera collection transmission line of electricity vision signal on shaft tower, through video server, it being sent back Surveillance center by transmission channel in real time in the mode of video flowing, intercepting contains the digital picture of conductor spacer from the transmission line of electricity video flowing that is sent to Surveillance center, employing carries out transmission line of electricity image target area location based on the Fast template matching algorithm of the field of direction and the conductor spacer coupling is located, to simplify the complexity of location algorithm, improve accuracy and the real-time of location, for the extraction of transmission line status information provides reference information; Concrete steps are as follows:

1) by being erected at the camera collection transmission line of electricity vision signal on shaft tower, intercepting contains the original image of conductor spacer from the transmission line of electricity vision signal;

2) intercepting conductor spacer template image to be positioned from original image;

3) to original image standardize successively, the field of direction is asked for, each calculates and image of zone marker is processed to consistance, obtain in zone marker image each to the zone of consistance value more than or equal to maximal value 96%;

4) marked region is carried out being undertaken by the mode of rank scanning the template matches of conductor spacer, the conductor spacer image that obtains take the 2nd step intercepting is as template, matching factor is take Matching sub-image top left corner pixel coordinate as index, the pixel coordinate of choosing maximum matching factor place is best match position, with the centre coordinate of the best match position place subimage centre spot as conductor spacer;

5) repeated for the 3rd, 4 steps, after repeatedly processing, mating, the coordinate of choosing probability of occurrence maximum in matching result is the conductor spacer anchor point.

Described normalization is original image to be carried out conventional pixel specification process, and the impact of removal of images gray difference is to ask for for the follow-up field of direction preprocessing process that carries out.

It is to adopt square gradient algorithm that the described field of direction is asked for, and adopts gradient vector and the deflection of image block method computed image sub-block, and then asks for integral image square gradient direction field.

Described each calculates and zone marker to consistance, respectively calculating to consistance numerical value the image direction field, take image each to consistance peaked 96% as threshold value, choose each and lock to the zone of consistance value more than or equal to this threshold value, and be labeled as the target area.

Described template matches is to adopt based on the histogram Fast template matching algorithm relevant with normalization, histogram is united the threshold value coupling and the normalization relevant matches combines, and reduces the complexity of mating calculating.Histogram associating threshold value matching algorithm is to adopt Euclidean distance and crossing function that template histogram and the histogram that mates subregion are carried out similarity to calculate, to improve real-time and the accuracy of algorithm.

The invention has the beneficial effects as follows that at first the transmission line of electricity image section to the conductor spacer place carries out target-region locating, simplifies the scope of coupling location scanning; Then adopt first histogram associating threshold value coupling, the method for the relevant optimum matching of rear normalization is carried out the conductor spacer centralized positioning; Choosing at last the coordinate that maximum probability occurs from the multiple bearing result is the best located point.The method has reduced the error rate in the position fixing process when having guaranteed the location real-time.

Description of drawings

Fig. 1 is the transmission line of electricity original image.

Fig. 2 is the image after the original image normalization is processed.

Fig. 3 is the direction field pattern of normalized images.

Fig. 4 is that each of transmission line of electricity image is to consistance figure.

Fig. 5 is the zone location figure as a result of transmission line of electricity image.

Fig. 6 is the Histogram Matching result queue figure of transmission line of electricity image.

Fig. 7 is the optimum matching positioning result figure of transmission line of electricity image.

Fig. 8 is that the coupling of transmission line of electricity image is located the three-dimensional table diagram.

Fig. 9 is the process flow diagram of transmission line of electricity images match location.

Embodiment

The invention provides a kind of method of utilizing video image processing technology to locate the transmission line of electricity conductor spacer, the method is a kind of new identification of transmission line of electricity conductor spacer and localization method based on the digital remote video image processing technology.The method is in order to identify accurately and locate the conductor spacer of transmission line of electricity, the digital picture that intercepts in the transmission line of electricity video flowing that is sent to Surveillance center is as research object, image is processed and identification by the digital picture that collects is carried out, extract transmission line of electricity conductor spacer center information, for the monitoring transmission line of electricity obtains sag, the running state parameter such as waves reference information is provided.Below in conjunction with accompanying drawing, the present invention is further described, and take the processing procedure of transmission line of electricity conductor spacer image as example.

One, Image Normalization

Fig. 1 is the original image (Fig. 1 top is fixed on conductor spacer on each electric wire, and is the external environment background image of camera collection beyond each electric wire) of transmission line of electricity conductor spacer.For the difference on the gradation of image that causes due to noise or the external environment of sensor (camera) self in the removal of images gatherer process, at first to the processing of standardizing of the original image of Fig. 1, difference on the removal of images gray scale, Fig. 2 is the image after standardizing.Normalization does not change the sharpness of image, just contrast and the gray scale of different target in image is adjusted on a fixing rank, for subsequent treatment provides a comparatively unified picture specification.The normalization formula is:

$G(i,j)=\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="0"\; label="M"\; filenames="HDA00002827157200022.png"\; state="\{\{state\}\}">M</figure-callout>}}_0+\sqrt{\frac{{\mathrm{VAR}}_0{(I(i,j)-M)}^2}{\mathrm{VAR}}},I(i,j)>M\\ M_0-\sqrt{\frac{\mathrm{VA}{R}_0{(I(i,j)-M)}^2}{\mathrm{VAR}}},I(i,j)\&le;M\end{array}\right.$

Wherein, I (i, j) is the gray-scale value of (i, j) coordinate points in original image, and M and VAR are respectively gray average and the variance of image I, M ₀And VAR ₀Be respectively the average of image and the expectation value of variance, G (i, j) is the rear image pixel of normalization.

Two, carry out the transmission line of electricity target-region locating by the field of direction

In order to simplify the computation complexity of location algorithm, adopt the target-region locating method based on the field of direction that transmission pressure zone in image is positioned.

The field of direction is the two dimensional surface field of Description Image grain direction and correspondence position thereof.Utilize the field of direction can obtain the partial structurtes characteristic of image, thus the realize target zone location.Can directly utilize the gradient fields of local data based on the method for gradient vector field, the statistical nature of comprehensive neighborhood generates the stable field of direction, can also assess the consistance that generates direction simultaneously.It is as follows that step is asked in square gradient direction field:

1) ask for the gradient vector [G of pixel (x, y) in image G _x(x, y), G _y(x, y)] ^T

2) piece (selecting w=8 herein) that image G is divided into a series of nonoverlapping sizes is w*w, the gradient vector of each piecemeal is:

${[G_{\mathrm{Bx}},G_{\mathrm{By}}]}^T=\left[\begin{array}{c}\underset{x=1}{\overset{W}{\mathrm{\&Sigma;}}}\underset{y=1}{\overset{W}{\mathrm{\&Sigma;}}}{G}_x^2(x,y)-G_y^2(x,y)\\ \underset{x=1}{\overset{W}{\mathrm{\&Sigma;}}}\underset{y=1}{\overset{W}{\mathrm{\&Sigma;}}}2G_x(x,y)G_y(x,y)\end{array}\right]$ ，

In formula, [G _Bx, G _By] ^TBe the gradient vector of piecemeal.

3) ask for piece deflection θ by gradient vector:

$\mathrm{\&theta;}=\frac{1}{2}\mathrm{\&pi;}+\frac{1}{2}\left\{\begin{array}{c}\arctan \left(\frac{G_{\mathrm{By}}}{G_{\mathrm{Bx}}}\right),G_{\mathrm{Bx}}\&GreaterEqual;0;\\ \arctan \left(\frac{G_{\mathrm{By}}}{G_{\mathrm{Bx}}}\right)+\mathrm{\&pi;},G_{\mathrm{Bx}}<0\&cap;G_{\mathrm{By}}\&GreaterEqual;0;\\ {\arctan }^{-1}\left(\frac{G_{\mathrm{By}}}{G_{\mathrm{Bx}}}\right)-\mathrm{\&pi;},G_{\mathrm{Bx}};0\&cap;G_{\mathrm{By}}<0.\end{array}\right.$

In piecemeal, the deflection of each pixel is identical, and θ ∈ [0, π).θ is carried out discretize, obtain corresponding direction number, thereby form the field of direction.

Consistance coh by the piece direction can weigh the reliability of piece direction, and the larger illustrated block inside gradient field of direction consistance of this value is better.[G in following formula _sx(x, y), G _sy(x, y)] ^TIt is square gradient vector of pixel in piece.

$\mathrm{coh}=\frac{\left|\underset{x=1}{\overset{W}{\mathrm{\&Sigma;}}}\underset{y=1}{\overset{W}{\mathrm{\&Sigma;}}}(G_{\mathrm{sx}}(x,y),G_{\mathrm{sy}}(x,y))\right|}{\underset{x=1}{\overset{W}{\mathrm{\&Sigma;}}}\underset{y=1}{\overset{W}{\mathrm{\&Sigma;}}}\left|(G_{\mathrm{sx}}(x,y),G_{\mathrm{sy}}(x,y))\right|}$

Square gradient vector is asked for according to the following formula: $\left[\begin{array}{c}{G}_{\mathrm{sx}}(x,y)\\ G_{\mathrm{sy}}(x,y)\end{array}\right]=\left[\begin{array}{c}{G^2}_x(x,y)-{G_y}^2(x,y))\\ 2G_x(x,y)G_y(x,y)\end{array}\right]$

In order to obtain the more specific location information of transmission pressure part, Fig. 2 travel direction field and each to be calculated to consistance, result is respectively as shown in Fig. 3,4.In Fig. 4, in the piece direction of selecting conductor part one pocket and image, the piece direction of other parts is carried out consistance and is calculated, and selecting the consistance value is the target area higher than the zone of maximal value 96%, the rower of going forward side by side notes, and result is as shown in Figure 5.

Three, employing is located conductor spacer based on the histogram Fast template matching algorithm relevant with normalization

On the basis of target-region locating, the present invention adopts based on the histogram Fast template matching algorithm relevant with normalization the conductor spacer center is positioned.

The Histogram Matching algorithm is reduced to one dimension with two-dimensional image information and is processed, and greatly reduces the complexity of calculating, but this algorithm accuracy is not high.For utilizing the Characteristic of Image Description Image, can be by the statistic histogram of gradation of image:

$p\left(r_k\right)=n_k/n,k=\mathrm{1,2},...,L$

Wherein, total number of image pixels is n, and the variation range of all grey scale pixel values is 1 ~ L, and gray-scale value is r _kNumber of pixels be n _k, p (r _k) be histogram.Usually adopt Euclidean distance function or histogram intersection function that histogram similarity is measured, definition is shown below respectively, P in formula _G(i), P _T(i) represent respectively the histogram of zone to be matched and template.

$M_E(G,T)=\sqrt{\underset{i=1}{\overset{L}{\mathrm{\&Sigma;}}}{[P_G\left(i\right)-P_T\left(i\right)]}^2}$

$Q_E=(G,T)=\frac{\underset{i=1}{\overset{L}{\mathrm{\&Sigma;}}}\min [P_G\left(i\right),P_T\left(i\right)]}{\underset{i=0}{\overset{L}{\mathrm{\&Sigma;}}}{P}_G\left(i\right)}$

Normalized Cross Correlation Algorithm is based on the classic algorithm in the Image Matching technology, and this algorithm accuracy is high, but calculated amount is large, is defined as follows:

$G_{\mathrm{NCC}}=\frac{\underset{(x,y)\&Element;W}{\mathrm{\&Sigma;}}G(x,y)*T(x,y)}{\sqrt{\underset{(x,y)\&Element;W}{\mathrm{\&Sigma;}}G{(x,y)}^2\underset{(x,y)\&Element;W}{\mathrm{\&Sigma;}}T{(x,y)}^2}}$

In order to take full advantage of the advantage of histogram and normalization relevant matches, overcome its defective, the method that the present invention adopts histogram and normalization relevant matches both to combine, set histogram associating threshold value optimum matching standard relevant with normalization, the transmission line of electricity conductor spacer is located fast and accurately.Algorithm steps is as follows:

1) read in original transmission line of electricity image, leave in buffer zone, comprise former figure each point pixel value, former figure width, height etc.;

2) intercepting conductor spacer image to be matched is matching template from input picture, calculates its histogram;

3) input picture is carried out target-region locating, the target area is carried out from left to right, search from top to bottom, and adopt respectively Euclidean distance and crossing function that template histogram and the histogram that mates subregion are carried out similarity and calculate;

4) determine the associating threshold alpha of Euclidean distance and crossing function=130 by repetition test, β=0.99 is selected greater than the pixel of the threshold value row labels of going forward side by side;

5) calculating of normalization relevant matches is carried out in image-region and the template image of mark, matching factor is take coupling subregion top left corner pixel coordinate as index, and the coordinate points of choosing the related coefficient maximum is rreturn value, is optimal match point;

6) characteristics axisymmetric according to conductor spacer, the central point of position matching subregion is the center of conductor spacer.

On the basis of target-region locating, histogram is carried out in the target area that marks in Fig. 5 slightly mate, return to the matching result in threshold value, as shown in Figure 6.The Histogram Matching result is carried out the normalization relevant matches, realize that finally the location of conductor spacer is as shown in square frame in Fig. 7.The three-dimensional table diagram of experimental result as shown in Figure 8, the three-dimensional table diagram shows that the optimum position, location is accurate and unique.

Fig. 9 is the process flow diagram of transmission line of electricity conductor spacer coupling location.

The method is in order to identify accurately and locate the conductor spacer of transmission line of electricity, the digital picture that intercepts in the transmission line of electricity video flowing that is sent to Surveillance center is as research object, image is processed and identification by the digital picture that collects is carried out, extract transmission line of electricity conductor spacer center information, for the monitoring transmission line of electricity obtains sag, the running state parameter such as waves reference information is provided.

Claims (5)

1. a method of utilizing video image processing technology to locate the transmission line of electricity conductor spacer, is characterized in that, the method is a kind of identification of transmission line of electricity conductor spacer and localization method based on the digital remote video image processing technology; By being erected at camera collection transmission line of electricity vision signal on shaft tower, through video server, it being sent back Surveillance center by transmission channel in real time in the mode of video flowing, intercepting contains the digital picture of conductor spacer from the transmission line of electricity video flowing that is sent to Surveillance center, employing carries out transmission line of electricity image target area location based on the Fast template matching algorithm of the field of direction and the conductor spacer coupling is located, to simplify the complexity of location algorithm, improve accuracy and the real-time of location, for the extraction of transmission line status information provides reference information; Concrete steps are as follows:

1) by being erected at the camera collection transmission line of electricity vision signal on shaft tower, intercepting contains the original image of conductor spacer from the transmission line of electricity vision signal;

2) intercepting conductor spacer template image to be positioned from original image;

3) to original image standardize successively, the field of direction is asked for, each calculates and image of zone marker is processed to consistance, obtain in zone marker image each to the zone of consistance more than or equal to maximal value 96%;

4) marked region is carried out being undertaken by the mode of rank scanning the template matches of conductor spacer, the conductor spacer image that obtains take the 2nd step intercepting is as template, matching factor is take the pixel coordinate of Matching sub-image as index, the pixel coordinate of choosing maximum matching factor place is best match position, with the centre coordinate of the best match position place subimage centre spot as conductor spacer;

5) repeated for the 3rd, 4 steps, after repeatedly processing, mating, the coordinate of choosing probability of occurrence maximum in matching result is the conductor spacer anchor point.

2. a kind of method of utilizing video image processing technology location transmission line of electricity conductor spacer according to claim 1, it is characterized in that, described normalization is original image to be carried out conventional pixel specification process, the impact of removal of images gray difference is to ask for for the follow-up field of direction preprocessing process that carries out.

3. a kind of method of utilizing video image processing technology location transmission line of electricity conductor spacer according to claim 1, it is characterized in that, it is to adopt square gradient algorithm that the described field of direction is asked for, gradient vector and the deflection of application image method of partition computed image sub-block, and then ask for integral image square gradient direction field.

4. a kind of method of utilizing video image processing technology location transmission line of electricity conductor spacer according to claim 1, it is characterized in that, described every consistance is calculated and zone marker, respectively calculating to consistance numerical value the image direction field, take image each to consistance peaked 96% as threshold value, choose each and lock to the zone of consistance value more than or equal to this threshold value, and be labeled as the target area.

5. a kind of method of utilizing video image processing technology location transmission line of electricity conductor spacer according to claim 1, it is characterized in that, described template matches is to adopt based on the histogram Fast template matching algorithm relevant with normalization, histogram is united threshold value coupling and the normalization relevant matches combines, reduce the complexity that coupling is calculated, adopt Euclidean distance and crossing function that template histogram and the histogram that mates subregion are carried out similarity and calculate, improve real-time and the accuracy of algorithm.

Images