JP2003004642A - Robot for observing floor material degradation degree and system for managing floor material by the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a robot for observing a degree of degradation of floor materials which is prevented from judgment variations and can efficiently judge by solving problems resulting from visual judgements, and a system for managing floor materials by the same. SOLUTION: An electronic camera or electronic scanner for imaging a floor is set to a vehicle. A vehicle position acquisition means is set for obtaining information on a vehicle position where the image is obtained. The robot for observing the degree of degradation of floor materials observes the degree of degradation of many floor materials set to the floor by relating images and the vehicle position information with each other.

Description

Detailed Description of the Invention

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for managing floor materials installed in offices and stores.

2. Description of the Related Art When floor materials such as carpets and tiles are placed in an office or a store, if the floor materials become old due to aging, they are wholly or partially replaced with new floor materials. . In the past, the timing of replacement with a new floor material was determined visually. The degree of deterioration of the floor material was judged with the naked eye, and the old and new floor materials were replaced at an appropriate timing.

Problems to be solved by the invention are apt to cause individual differences in visual judgment. Depending on the person who makes the judgment, the degree of deterioration of the flooring material considered to need replacement tends to vary. Further, when a large amount of floor material is arranged, there is a drawback that the visual judgment requires a lot of time and labor. The object of the present invention is to eliminate the drawbacks caused by visual judgment,
It is an object of the present invention to provide a floor material deterioration degree observing robot capable of making judgments with little variation and capable of making efficient judgments, and a floor material management system using the same.

The solution means of the present invention is exemplified as follows. (1) An electronic camera or an electronic scanner for taking an image of the floor surface is provided in the vehicle, and vehicle position obtaining means for obtaining the vehicle position information by taking the image is provided, and the image and the vehicle position information are associated with each other to obtain the floor surface. A floor material deterioration degree observing robot, which is capable of observing the degree of deterioration of a large number of floor materials installed in a building. (2) Floor material management characterized by storing the floor material deterioration degree obtained by the floor material deterioration degree observation robot in a storage device of a computer and managing the floor material by comparing it with the floor material arrangement data. system.

BEST MODE FOR CARRYING OUT THE INVENTION A preferred embodiment of the present invention will be described. The degree of deterioration (wear) of the floor material is estimated by using a vehicle (moving carriage) equipped with an electronic camera or an electronic scanner at a store or office, and the data is stored in a storage device of a computer to arrange the floor material. Relocate, remove, and recycle flooring materials against the data. Preferably, flooring materials such as carpet tiles are produced and those flooring materials are rented to offices or stores. Then, the installation location of each floor material is managed on the CAD drawing of the office or store. The floor material deterioration degree observation robot monitors the usage status of many floor materials installed on the floor in the room and manages each standardized floor material (for example, 50 cm square) using a database linked to a computer. To do. The degree of wear of the floor material is expressed by an index (index) such as 0 to 100, and the floor material is exchanged with a new floor material or rearranged so that it can be effectively used. In order to evaluate the degree of wear of each floor, we use video data from electronic cameras and scanners installed in vehicles moving in stores and offices.

The preferred embodiments of the present invention will be described below. FIG. 1 shows an example of a vehicle equipped with an electronic camera. FIG. 2 shows a control system configuration diagram of a vehicle equipped with an electronic camera. In FIG. 1, a vehicle 1 has an electronic camera 3 for capturing an image of a floor surface 2, a display 4 for capturing the image, a battery 5, and wheels 6. As shown in FIG. 2, the vehicle 1 further includes a controller 7 that sends a shutter control timing signal to the camera 3. The controller 7 receives the information from the rotation sensor 8, calculates the position coordinates of itself, and controls the motor 9. The video captured by the camera 3 is recorded on the hard disk 10 and is then carried to an FPC (floor planning center) for analysis processing. Alternatively, while capturing the image, it is simultaneously transmitted to the FPC by radio waves. The method of using the electronic camera-equipped vehicle 1 in the room is as follows. The electronic camera 3 photographs the floor surface 2. Using this electronic camera 3, the degree of wear and deterioration of the floor surface 2 is evaluated according to the following relationship. Movement of Electronic Camera 3-> Position in Image-> Evaluation of State of Floor 2 When the electronic camera 3 moves, it is necessary to accurately grasp the position of the electronic camera 3. Therefore, as shown in FIG. 3, a mark 11 for calibrating the start position is provided in a part of the room.
When the moving vehicle 1 starts running, this mark 1
Make sure to check 1. The current position of the moving vehicle 1 is obtained by integrating the rotation amount of the wheels 6 and the rotation angle of the steering wheel. As a result, the position of the vehicle 1 is always calculated. An image is taken while the vehicle 1 is moving, and a still image is taken by a sample (depending on the moving speed and the required resolution) at regular time intervals and registered in the hard disk 16 for each position coordinate of the vehicle 1 or directly by radio waves. Transfer to FPC. The FPC processes information received by radio waves or information stored in the hard disk 10. First, after confirming that the entire shooting has been completed, as shown in FIG. 4, the respective still images 12 are arranged on the position coordinates,
The overlapping part is removed and the whole image is processed into one image data. An electronic scanner may be used instead of the electronic camera 3. In order to evaluate each floor material, image data of a scanner installed in a vehicle 1 moving to a store or office is used. 6 to 7 show an example of a mobile scanner. Scanner is linear
The linear lens 61 converts the image of the floor surface 2 into a digitized linear image. Illuminating light is applied to capture deterioration and wear of the floor surface 2 as an image. In this case, the line scanner 62 is used as a light source instead of the electronic camera. By using the scanner, detailed data of the floor surface 2 can be obtained from a very short distance.
Data can be collected by licking the floor.
The data input from the line scanner is linear data, which is accumulated according to the traveling speed to form one piece of video data. Since this method can capture a detailed image of the floor material, it is a method suitable for constructing a resource recycling system by evaluating deterioration and wear of the floor material. The scanner also has a controller 64, calculates position coordinates of itself, and controls the motor 63. The captured video is recorded on the hard disk, and later transported to an FPC (Floor Planning Center) for analysis processing. Alternatively, while capturing the image, it is simultaneously transmitted to the FPC by radio waves. Humans may push it by hand to propel a small vehicle,
It is also possible to drive automatically. In this case, a steering device is attached to the front wheels so that the steering wheel can be turned to the left and right while moving. The rear wheels are driven by the motor 63. When the indoor CAD data is received from the FPC (Floor Planning Center), the position of the obstacle can be transmitted to the vehicle in advance, so if you approach the obstacle, slow down the speed and use the steering device to select an appropriate traveling route. can do. The electronic camera moving vehicle and the linear scanner moving vehicle described above can be realized in combination with the function of the automatic cleaning robot. In this case, when the automatic cleaning robot moves while cleaning the floor surface, it records the image of the corresponding floor surface each time cleaning is completed, and uses this information to perform floor maintenance. You can In addition to this, there is a method of observing the condition of the floor material using the image. One is to put fluorescent paint on the floor or paint it, and illuminate it with a black light to create an image. In this way, wear and stains are hard to see under ordinary light sources during business hours in stores, but you can clearly see the fluorescent part by shining a black light (ultraviolet light source) at night. When the floor material wears down, the fluorescent paint wears down, and the change can be captured in the image. Alternatively, a material that is sensitive to infrared rays may be included in the flooring material or painted, and the infrared rays may be irradiated to record the image on the camera. In this case, the image seen by infrared rays changes depending on the degree of wear of the floor material, and the degree of wear is displayed on the infrared camera. In any of the electronic cameras and linear scanners described above, using fluorescent paint and black light, infrared compatible images, etc., any of the embodiments,
Once the image data is obtained, the subsequent processing is the same. Figure 6 is an overall view of a linear scanner mobile vehicle.
FIG. 7 is a plan view. A method of using the scanner type vehicle of FIGS. 6 to 7 will be described in more detail. A moving linear scanner captures the floor surface. Using this linear scanner, the degree of wear and deterioration of the floor surface is evaluated according to the following relationship. Movement of linear scanner-> Position in image-> Evaluation of floor condition When the linear scanner moves, it is necessary to accurately grasp the position of the linear scanner. Therefore, a mark for calibrating the start position is attached to a part of the room, and this mark is confirmed when the moving vehicle starts traveling. The current position of the moving vehicle is obtained by integrating the rotation amount of the wheel and the rotation angle of the steering wheel. With this, the position of the vehicle is constantly calculated. An image is taken while moving, and a still image is taken by a sample (depending on the moving speed and the required resolution) at regular time intervals and registered in the hard disk for each position coordinate of the vehicle, or directly transferred to the FPC by radio waves. The FPC processes the information received by radio waves or the information stored in the hard disk. First, after confirming that the whole shooting is completed, the respective still images are arranged on the position coordinates, and as shown in FIG. 4, the overlapping portions are removed and the whole becomes one piece of video data. To process. Information on the floor surface of the target room is extracted from the CAD data, divided into grids of floor material size, and this is overlaid on the video data to determine the state X (k) of each floor material k.
Correspond to. Next, the state X of each floor material k is calculated. First, the brightness information Y is obtained from the data of the three primary colors R, G, B of the pixel of the video data. Alternatively, from the beginning, the scanner should capture the brightness as black and white gray data only. The conversion method from the data of the three primary colors of red, green, and blue to the lightness Y is performed as follows. Y = 299 * R + 587 * G + 114 * B Next, the value of Y is examined over the entire area of one flooring material, and a brightness histogram is created. The histogram is
The brightness is divided into several stages to represent the frequency with which the Y value of each pixel occurs in each division. First, a brightness histogram is created using the obtained image as brightness information.
As shown in FIG. 5, the image histogram shows the brightness Y of each pixel of the image on the horizontal axis and the frequency of appearance on the vertical axis.
In the illustrated example, the horizontal axis represents the luminance Y in 256 levels, and the vertical axis represents the frequency of occurrence of pixels having the luminance Y. In order to simplify information processing, it is convenient to divide the horizontal axis into 10 to 20 stages and express them collectively as shown in FIG. Since the image histogram can simply express the characteristics of the image, the present invention can be effectively utilized in various ways. By comparing the image histograms (H) of the same floor material at appropriate intervals, it is possible to evaluate the degree of deterioration or wear of the floor material. Since the brightness generally decreases as the deterioration progresses, the frequency of occurrence of a section near the origin (from the left) in the horizontal axis direction increases. For flooring, there are patterns with colors and patterns for design reasons.
Even in such a case, if the change amount of the video data is taken out and the histogram is taken, the information of the color and the pattern which is initially contained disappears as the change amount and can be treated in the same manner. Further, depending on the conditions, it may be possible that stains having a light color tone are generated on a dark floor material. For example, when milk is spilled on a black floor material. In this case, the degree of wear on the dark color side is low. If the brightness of the floor material is known in advance, the bright side and the dark side of the histogram in the following description can be understood in reverse, so a detailed description will be omitted. H (t, k, n) = H (t, y1, y2, y3, ... y
N, k) H (t, k, n): Image histogram of floor material k at time t yn: Frequency of lightness of the nth section (n = 1, 2, ... N) Change of histogram H at time t, t + 1 Is as follows. ΔH (t, k, n) = H (t + 1, k, n) −H (t,
k, n) Actually, the value of each section is obtained by subtraction. H indicates the state of the floor material. It is considered that the degree of deterioration or wear appears in the change of the histogram H of the brightness information. The larger the value of ΔH, the larger the load is expressed. The flooring material k placed at the position i receives a load of ΔH. Since the floor material k is reused and the position i changes, it is necessary to remember that the floor material k was at the position i at the time T. s (T, k) = it-1 <T <= t H (t, k, n) indicating the state of the flooring material k is a matrix having only the number of histogram divisions n, and is not a scalar quantity.
It is not suitable for simply comparing and evaluating the degree of deterioration and wear of floor materials. Therefore, the following indicators are created. The difference from the histogram at t = 0 is squared and evaluated for each section n: division number = 1, 2, 3, 4, ... N X (t, k) is a scalar quantity, and if 0, the degree of deterioration (wear degree) is the most severe, and if large, the degree of deterioration (wear degree) ) Becomes smaller. Further, it is conceivable that the degree of soiling will not be uniformly reduced by cleaning or repairing the floor material, but that the state of being reduced to a certain point will be recovered. In such a case, the state at that time is regarded as the initial state. Specifically, H (t, k,
n) may be reset as an initial value. Now, if X (t, k) indicating the state of the floor material k becomes smaller than a certain level, it is judged that the degree of the wear exceeds the limit,
Stop using the floor material, take it back to the factory, and put it in the recycling process. Calculate which flooring will be placed in the empty space created here. There are various possible methods for this calculation, but one example is to minimize the cost of replacing floor materials and the number of floor materials actually replaced. Removal of flooring material with severe wear Removal of flooring material with severe wear is replaced when the level falls below a specified level E (for example, 10-15 as an index). Therefore, the position j where X (t, k) <= E is found. The flooring material at the position j that satisfies this condition is removed, returned to the factory, and placed in the recycling process. A floor material in another position or a new input floor material is applied to the vacant position j. The method of ascertaining which flooring to apply is important. Evaluation function for floor material replacement Since many combinations are required to replace floor materials, an evaluation function is used to make the plan quantitative. The floor material m to be applied is either new (x = 100 highest level) or the floor material already used. The cost evaluation function C of the construction required to apply this floor material can be expressed in various ways, and the following examples can be considered. (1) Floor material replacement distance cost Cost function C is distance d to floor material m at position j
Expressed as (j, m). The cost will be reduced if floor materials that are as close as possible are applied. Since the new floor material is brought in from a distance, the distance cost becomes large. In this case, as an example of the solution that minimizes C, the new floor material is replaced with the level E floor material near the doorway, and the level E floor material in the back is relatively close to the surrounding area. Replaced with undamaged flooring. (2) The prediction calculation is performed on the assumption that the floor material k is at the position i by the time (t + 1) when the fixed period of the next exchange floor material elapses and that the same load as this time will occur. x (t + 1, k) = 2 * x (t, k) -x (t-1,
i) With this, the state of the floor material k is calculated. The number of E-level removal floor materials generated at this time is N. If N is large, the cost increases accordingly, so this is added to the cost function. At this time, the load x (t, i) -x (t-1,
Since i) can fluctuate significantly due to a temporary decrease, the load for a certain period (eg t-3 as t,
It is also effective to use a moving average of (t-2, t-1, t). (3) Avoid concentration of severely deteriorated parts If the severely deteriorated flooring is concentrated only on a part, the impression is deteriorated both visually and psychologically. Therefore, the degree of deterioration should not be concentrated in advance. The whole is divided into several zones p (p = 1, 2, 3 ...), the average value Xp of the respective wear states is calculated, and the variance from the average X of the whole wear states is evaluated. That is, the overall average X is as follows. The average Xp of the floor material included in the zone p is as follows. Xp = Σx (t + 1, i) / Mz (Mz is a set of positions i belonging to the zone p) (Calculate for i belonging to P) How much the zone p deviates from the whole can be expressed by Jp. Since Jp represents the degree of deviation from the whole, it is considered that if Jp is large, the deteriorated portion is not averaged and the whole becomes uneven and the cost is high. Jp
Let J be the sum of all p. The cost evaluation function I generated in the above process is synthesized as follows. The combination that minimizes this cost evaluation function is calculated. I = C + w ₁ xN + w ₂ xJ-> Minimization
(W ₁ and w ₂ are weighting factors) There are various calculation methods, but the simplest method is to enumerate all combinations and calculate I, and take out the case where I becomes the smallest. Thereby, the relocation plan of each floor material can be determined. This layout recalculation is a kind of optimization problem, and mathematical methods such as dynamic programming, integer programming, and branch-and-bound combinatorial calculation can be applied. FIG. 8 shows a brief flow chart of the above process. Now, finally, floor material relocation calculation is executed by communicating with a computer in an FPC (Floor Planning Center). The calculation result is displayed on the CAD drawing in the FPC database so that the design can be examined. Check whether the calculation results of the cost function meet the conditions of flooring layout design. If they do not match, the above calculation is performed under the design constraint conditions to find the solution.

According to the present invention, in a store or office, image information of a floor material is captured by an electronic camera or an electronic scanner to estimate the degree of deterioration or the degree of wear of the floor material, and the data of the state of the floor material is acquired. Can be stored in the storage device of the computer of the FPC, and the floor material can be updated, rearranged, removed, or recycled by collating with the floor material layout data in the store or office. Therefore, floor material management can be sold as a service,
It contributes to the efficient use of resources and ensures that flooring users always receive a clean and high standard of service.

[Brief description of drawings]

FIG. 1 shows an example of a floor material deterioration degree observing robot according to the present invention.

2 shows an example of a system of the robot of FIG.

3 shows an example of start calibration in the system of the robot of FIG.

4 shows an example of synthesizing video data by the system of the robot of FIG.

FIG. 5 shows an example of a histogram of image data.

FIG. 6 shows an example of a mobile scanner type vehicle according to the invention.

FIG. 7 is a plan view of the vehicle shown in FIG.

FIG. 8 shows an example of a flowchart of an electronic camera type flooring management system according to the present invention.

[Explanation of symbols]

1 vehicle 2 floor 3 electronic cameras 4 display 5 battery 6 wheels 7 controller 8 rotation sensor 9 motors 10 hard disk 61 linear lens 62 linear light source 63 motor 65 line sensor

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) // G01B 21/00 G01B 21/00 D (72) Inventor Hozumi Yagioka Hozumi Yagioka Tokyo 3rd 2nd, Chidanda, Tokyo No. 5 Toshin 24 Kudan Bldg. 6F In-house Co., Ltd. (72) Inventor Masaaki Oe 3-11-13 Iwamotocho, Chiyoda-ku, Tokyo Tajima in stock company (72) Inventor Haruki Masakiya 2-1-6 Higashi-Nihonbashi, Chuo-ku, Tokyo F-term in Institute for System Technology (Reference) 2F069 AA03 BB40 GG04 HH30 NN06 NN25 2G051 AA90 AB02 AB07 AC15 AC16 AC19 AC21 BA05 BA06 BA20 BB09 CA03 CA04 EA14 EA16 EA17 EC02

Claims (2)

[Claims] 1. An electronic camera or an electronic scanner for taking an image of a floor surface is provided in a vehicle, and a vehicle position obtaining means for obtaining information of the vehicle position where the image is taken is provided, and the image and the vehicle position information are associated with each other. , A floor material deterioration degree observation robot characterized by observing the degree of deterioration of a large number of floor materials installed on the floor surface. 2. The floor material deterioration degree obtained by the floor material deterioration degree observation robot according to claim 1 is stored in a storage device of a computer, and the floor material is managed by comparing it with the floor material arrangement data. A characteristic floor material management system.