JPH01123374A - Crack image data processing method - Google Patents

Abstract

PURPOSE:To remarkably accelerate data processing speed by preventing the extraction of a segment from being performed in all areas by employing a tracking processing, and performing the extraction processing of a line segment and a tracking direction decision processing in parallel. CONSTITUTION:A tracking direction is decided based on position relation between a slit area decided that a crack exists at this time and the split area decided that the crack exists at a preceding time, and also, the tracking direction decision processing is performed in parallel with the extraction processing of the line segment. In other words it is not that the tracking direction is decided by a line direction obtained from the extracted result of the line segment, but it is decided based on the position relation between two split areas where the existence of the crack being obtained as the result of the decision of the presence/absence of the crack is confirmed, therefore, it is possible to execute the extraction processing of the line segment and the tracking direction decision processing of parallel, thereby, it is possible to shorten the processing time of data compared with a device in which all processings are executed in series.

Description

[Detailed description of the invention] [Industrial application field] This invention recognizes crack image data obtained by road surface measurement, extracts line segments corresponding to cracks, and outputs analysis results such as crack occurrence rate. Regarding the efficiency of the series of processing up to this point.

[Conventional technology]

There are three types of road surface properties that can be used to evaluate road damage: ruts, longitudinal unevenness, and cracks. Various proposals have been made to automate these measurements, which were conventionally done by visual inspection.

These automated technologies employ a slit camera method (Japanese Patent Publication No. 61-281915) and a laser light receiving method (Japanese Patent Application No. 58-229563, Japanese Patent Application No. 59/1989) as measurement methods.
-233923) has become mainstream, and an imaging camera, a laser oscillator, a light receiving element, etc. are mounted on a vehicle, and the measuring vehicle runs on the road to automatically measure road surface image data. In these proposals, the road surface image data obtained during the road surface needle 71pt is recorded on a data recorder such as a video tape recorder (VTR), and then the road surface is evaluated by analyzing this recorded data. That's what I do.

By the way, when analyzing such recorded data, conventionally, road surface evaluation is performed by human interaction with a monitor display screen. In other words, taking the laser method as an example, a videotape on which road surface image data is recorded is displayed on a monitor via a playback device, and a grid-like mesh is applied to the display screen to divide the display screen into multiple square areas. A human visually judges the presence or absence of cracks in each of these divided areas, and the human enters the judgment result on a grid-shaped recording sheet with partition lines, for example, as Q, x, etc. Finally, evaluation data such as the crack occurrence rate (-number of meshes with cracks/total number of meshes) is calculated with reference to the recorded results.

[Problem that the invention seeks to solve]

In this way, in the conventional method, crack data included in road surface image data is analyzed and processed by humans visually judging the monitor display screen.
This method had problems such as slow data processing speed and erroneous judgments caused by visual confirmation.

This invention was made in view of these circumstances, and provides a crack image data processing method that speeds up and improves the accuracy of data processing by automating the analysis of crack image data using a computer. This is what we are trying to provide.

[Means and effects for solving the problem] Therefore, in the present invention, an image memory in which road surface data is stored is divided into a plurality of regions, and cracks are detected in some predetermined divided regions among these plurality of divided regions. The presence or absence of data is determined separately for each area, and the area in which it is determined that a crack exists is selected as the tracking starting point, and the following tracking processing (tracking direction determination processing, line The segment extraction process and the crack presence/absence determination process are repeated for each starting point.

At the time of starting tracking, first, the next tracking direction is determined based on the position of the starting point area, and it is determined whether or not a crack exists in the divided area adjacent to the starting point area on the subsequently determined tracking direction, If the existence is confirmed, line segments are extracted in the area. From then on, by repeating the tracking direction determination process, crack presence/absence determination process, and line segment extraction process in the same manner, cracks included in the predetermined area are extracted as a combination of line segments. In the process, the tracking direction is determined based on the positional relationship between the divided area where it was currently determined that a crack exists and the divided area where it was previously determined that a crack existed, and this tracking direction determination process is combined with the line segment extraction process. Try to do this in parallel. In other words, in this process, the next tracking direction is not determined by the line direction obtained from the line segment extraction results, but by the positional relationship between the two divided regions in which the presence of cracks has been confirmed, which is obtained from the crack presence/absence determination results. Therefore, the line segment extraction process and the tracking direction determination process can be executed in parallel, reducing data processing time compared to a system in which all processes are executed in series. can.

〔Example〕

Hereinafter, the present invention will be described in detail according to embodiments shown in the accompanying drawings.

FIG. 2 shows an example of a configuration for implementing the present invention. In this case, road surface image data measured by a road surface measuring vehicle using a laser method is recorded on the videotape 1 shown in FIG. The conversion module 2 converts the road image data recorded on the videotape 1 into lff1mX1a+, for example.
Each m pixel is quantized into 8 bits (256 gradations) and written onto the magnetic tape 3. The crack data obtained using the laser method is expressed as a gradation image from white to black due to the shadow effect, where cracks are black (gradation 0) and road surfaces without cracks are black (gradation 0). Gray, white center lines, etc. are expressed as white (gradation 255).

The computer 4 has an image memory that stores crack data transferred frame by frame from the magnetic tape 3, processes this crack data in the manner described below, and outputs the processing results to the printer 5. , monitor 6
display it on top.

In this case, line segment extraction processing (described later) included in the data processing procedure is executed by a dedicated processor 7.

Hereinafter, an example of the data processing procedure according to the present invention will be explained in order with reference to the flowchart shown in FIG. 1 and the like.

The computer 4 transfers the road crack data recorded on the magnetic tape 3 to the image memory in the computer in frame units (for example, 512 x 512 pixels) (step 100), and executes the following analysis process on the transferred image data. do.

First, the computer 4 calculates one frame shown in FIG. 3, and uses the average value D1 to calculate a threshold value δ for determining the presence or absence of a crack to be used in the next starting point search process (step 110
).

Next, the computer 4 processes the transferred image data for one frame in 16×16 format, for example, as shown in FIG.
square slits, and these slits magnify the image by 16X.
Divide into 16 square slit areas (step 120)
. In this case, one slit area includes image data of 32×32 pixels.

The computer 4 first performs a "start point search" for cracks on the thus divided image data. This starting point search is for searching the starting point for the next tracking process (described below), and in this starting point searching process,
Rather than checking for cracks in all slit areas, check for cracks only in the slit areas where the arrows ■, ■, ■, ■ in the TS3 diagram intersect, in the direction of the arrows and in the order of ■, ■, ■, ■. Determine the presence or absence of. The reason why the starting point search area is partially restricted in this way is because cracks usually have a certain length, and the cracks correspond to the arrows ■, ■, ■, and ■ in Figure 3. If the area is covered with a net, the starting point can be detected in any of the slit areas.

Cracks are determined as follows. That is,
The computer 4 calculates a histogram consisting of the relationship between the gradation level and the number of pixels as shown in FIG.
The number N of pixels existing below (corresponding to the hatched area in FIG. 4) is determined for each slit area. Then, the computer 4 determines that there is a crack when the number N is greater than or equal to a predetermined set number Na, and determines that there is no crack when N<Na (step 130).

This idea is because the cracks cross the inside of the slit as shown in Figure 5(a), so when the cracks exist within the slit area (32x32),
This is based on the reason that there are pixels with density data close to 0 (corresponding to black) that are at least a predetermined value Na (for example, 36) or more. Therefore, in this embodiment, a histogram within the slit is created and this histogram is By performing threshold processing using a threshold value δ determined from the average value (or median value), etc. within the frame, the number N of pixels less than or equal to the threshold value δ is determined, and this is compared with a predetermined value Na. This is how we judge cracks to be nothing. - Next, the computer 4 performs line segment extraction processing in each slit region determined to have a starting point (crack) by the above-described starting point search processing (step 140). That is, in this line segment extraction process, if a crack pattern LP as shown in FIG. 45(a) exists in the slit area, it is extracted as a rectangular line as shown in FIG. The line width W is extracted as a segment pattern SG, and the extraction result is shown in the same figure (c).
One line length, one line direction θ, coordinates of line segment (point P. - P
3), the coordinates of the ridge line SL of the line segment SG (identified by the coordinates of points Qo and Q+), and the like.

In this case, the linear pattern recognition method disclosed in Japanese Patent Application No. 61-12490 is used for the above-mentioned line segment extraction process. Let me explain.

According to this recognition method, image data is divided into a plurality of (4×4 in this case) square slit regions as in FIG. Figure (a).

(b)). In this case, the projection waveform values S and S are set to be the average of the density data of the pixel column (n') in the projection direction.

Next, the slit is sequentially rotated by a predetermined angle in the range of 0 to 90 degrees with the center of each slit area as the rotation center, and a projected waveform is obtained for each rotation angle.

The method to obtain this rotated slit projection waveform is as follows:
There are two methods: one is to fix the image and rotate the slit, and the other is to fix the slit and rotate the image.

If the projected waveform is obtained for each rotation angle in this way, the projected waveform in the direction along the linear pattern LP will be obtained when rotated by a certain predetermined angle θ, as shown in FIG. 6(c). The peak value P becomes the maximum, and from this, the linear pattern existence direction θ can be determined.

Next, the line width W of the linear pattern is determined by the length W of the peak cut line when the projected waveform is subjected to threshold processing using a predetermined threshold value "hd" as shown in FIG. 6(c), for example. shall be.

Next, the length of the linear pattern is determined by limiting the slit area in the width direction only to the already obtained crack width W, as shown in FIG. The line length included in the slit area is determined by performing threshold processing using a certain threshold value Thd2. This is because with a simple square slit, the peak of the line clearly appears in the waveform along the line, but the peak contrast in the waveform in the width direction becomes low.

By performing such projection waveform analysis, it is possible to recognize the linear pattern included in one slit area as a rectangular pattern with known width W1 length L1 direction θ, etc., as shown in FIG. 6(e). can. As mentioned above, this is called a line segment.

When such processing is performed for each mesh region in FIG. 6(a), a linear pattern can be extracted for each mesh as a line segment whose width W and length direction θ are known.

In other words, in this starting point search process, the arrows ■, ■
By repeating the process of determining the presence or absence of a starting point and the line segment extraction process in the order of , ■, ■, a tracing starting point in the frame is selected, and a line segment is extracted from these selected tracing starting points. There is.

Then, the computer 4 assigns serial numbers to the slit regions selected by such a starting point search process (step 160), and executes the following "tracking process" in the order of these numbers.

First, for the starting point No. 1, the next tracking direction is determined based on the position of this starting point (step 170). That is, when the tracking starting point is at the position shown in FIG. 3(a), the adjacent right three slit areas are determined to be the next tracking area, as shown in FIG. 7(a). Similarly, the tracking starting points are (b), (c), (d), (e), (f),
(g), (h), the next tracking direction is shown in Fig. 7 (b), (c), (d), (e), (f), (g), (
Determine each as shown in h). In this way, in tracking when the tracking light is used as the tracking starting point, the tracking direction is specified only by the position of the tracking starting point within the frame.

Next, the computer 4 determines whether or not a crack actually exists in the tracking slit area determined in this way, using the determination method using the histogram shown in FIG. ).

If a crack is detected in this determination, the computer 4 operates the line segment extraction processor 7 to extract a line segment in the slit area determined to have a crack using the aforementioned projection process (step 210).

Further, in parallel with this line segment extraction process, the computer 4 performs a process of determining the next tracking direction (step 200). However, in this case, the tracking direction is determined based on the positional relationship between the previous slit that was determined to have a crack and the slit from the time before the previous one. That is, FIGS. 8(a) and (b)
As shown in , if tracking is performed directly horizontally (two slits determined to be cracked are adjacent on the left and right), the next tracking is performed for the three slits on the right or left, and the eighth As shown in Figures (c) and (d), if tracking has been performed directly below or above (when two slits are adjacent above and below), the next tracking is performed for the 3 slits below or above. , as shown in Fig. 8(e) to (h), when tracking has been performed diagonally (when two slits are diagonally adjacent), the following is true for the three slits shown centered on the diagonal slit: tracking.

By the way, one possible method for determining the tracing direction is to determine it based on the line direction θ of the line segment information, but in this case, the line segment extraction result cannot be obtained, and the next tracing direction is I can't decide. Therefore, the processing in this case is always a series.

However, in the method of the present embodiment described above, the tracking direction is determined based on the positional relationship between the current and previous slits that were determined to be cracked. can be determined. Therefore, it is no longer necessary to perform this tracing direction determination process after the line segment extraction process, and these tracing direction determination processes and line segment extraction processes can be executed in parallel after determining the presence or absence of cracks, improving the efficiency of data processing. It is possible to increase the speed.

As shown in FIG. 9, if it is determined that there are cracks in two or more slit areas in one tracking, it is determined that there is a "branch" (step 190), and the same tracking as described above is first performed in the direction of A. When the tracking is completed, the next tracking is performed in the direction of B.

The computer 4 repeatedly executes such tracking processing until the tracking area reaches the end area of one frame or there are no cracks in the tracking direction, thereby performing tracking and tracking starting from the starting point of No. 1. Executes line segment extraction processing. FIG. 10 shows a conceptual flow of such a series of tracking processing. When this tracing is completed (step 220), the computer 4 extracts the extracted line segment information (θ, W, L, segment coordinates, ridge line coordinates, etc.) and slit connection information (line segments are connected between adjacent slits). (information indicating whether or not)
etc. in memory (step 230), then fire N
The same tracing and line segment extraction processing as described above is executed for the starting point of O,2 (step 240). When the tracking and extraction processing for all starting points is completed (step 250), the computer 4 uses the line segment information, slit connection information, etc. to
Calculate evaluation data such as the incidence of cracks existing on the road surface and the length of cracks (step 260
). In this way, after the computer 4 finishes data processing for the frame, it transfers the road surface data of the next frame from the magnetic tape 3 again (step 270).
), and thereafter data processing in the next frame is performed in the same manner as described above.

FIG. 11 shows an example of such tracking and line segment extraction processing; in this case, No. 1 and No.
, 2, the tracking and extraction process is executed from two starting points.

In this way, in this embodiment, line segment information corresponding to a cracked image is obtained by executing a series of processes as shown in FIG. 10, rather than performing line segment extraction processing in all slit areas. As a result, the number of times line segments are extracted can be significantly reduced, and data processing speed can be significantly increased.

Furthermore, since the tracking direction is determined based on the positional relationship between the two tracking slits and the tracking direction determination process and line segment extraction process are parallelized, efficient processing can be achieved and processing time can be shortened.

〔Effect of the invention〕

As explained above, according to the present invention, the entire crack data extraction process is automated, so that erroneous judgments can be reduced, the recognition rate can be improved, and costs can be reduced by reducing the number of personnel. In addition, by adopting tracking processing, segment extraction is not performed in the entire area, and line segment extraction processing and tracking direction determination processing are parallelized, resulting in a significant increase in data processing speed. Can be done.

[Brief explanation of the drawing]

FIG. 1 is a flowchart showing an embodiment of the present invention, FIG. 2 is a block diagram showing a configuration example for carrying out the invention, FIG. 3 is an explanatory diagram for explaining an example of a starting point search process, and FIG. Figure 4 is a graph for explaining the method of determining the presence or absence of cracks, Figure 5 is a diagram for explaining the line segment extraction process, Figure 6 is a diagram for explaining Figure 5 in more detail, Figure 7 is a diagram for explaining the process for determining the tracking direction in the origin area, Figure 8 is a diagram for explaining the process for determining the tracking direction in the normal area, and Figure 9 is an example of branching. FIG. 10 is a diagram conceptually showing the flow of a series of processes according to the present invention, and FIG. 11 is a diagram showing an example of the extraction result of line segments in one frame. 1...Video tape, 2...Conversion module, 3.
...Magnetic tape, 4...Computer, 5...Printer, 6...Monitor, 7...Line segment extraction processor, LP...Crack pattern, SG...Line segment. Figure 1 Figure 2 Figure 4 (c) (d), 7△, Figure 6 (α) (b) (c
)(d)(e)(
f) (!; l) , (h) Figure 7 (a) (b) (e)
(f) (g) (h) Figure 8 Giant Association Figure 10

Claims (1)

[Scope of Claims] A crack image data processing method in which imaged road surface data is stored in an image memory in predetermined units, and line segments corresponding to crack data are extracted from the stored road surface data, wherein the image memory is Divide into regions, determine whether or not crack data exists in some predetermined divided regions among these multiple divided regions, and select the region where it is determined that cracks exist by this determination as the tracking starting point. The crack image data processing is characterized in that line segments corresponding to the cracks are extracted from the selected starting point area, and the following tracking process is performed for each starting point using the selected starting point area as a starting point. Method. (a) Determine the next tracking direction based on the position of the starting point area. (b) Determine whether crack data exists in the adjacent divided area in the determined tracking direction. (c) Extracting line segments is performed in the area where it is determined that a crack exists. (d) A process of determining the next tracking direction based on the positional relationship between the starting point area and the divided area currently determined to have a crack is performed in parallel with the line segment extraction process. (e) Determine whether or not crack data exists in the adjacent divided region in the determined tracking direction, perform line segment extraction processing in the region where it is determined that cracks exist, and if cracks exist, this time A process of determining the next tracking direction based on the positional relationship between the determined divided area and the divided area previously determined to have cracks is performed in parallel with the line segment extraction process, and these determinations, extractions, and directions are performed. A series of processes consisting of determination are repeatedly executed until tracking for the starting point is completed.