JPH0470533A - Oil-leakage detecting method - Google Patents

Abstract

PURPOSE:To make it possible to ensure the detection of oil leakage even for a moving object to be detected by forming the image of a region to be measured, extracting the tinted color, and judging the oil leakage when the computed value of the extracted area is smaller than the computed value of the previous area. CONSTITUTION:An image processing device 8 averaged five inputted images and forms a reference image. Then, the red color is extracted from the reference image. Thereafter, the extracted red color is binary-coded by a discriminating analysis method. Noise removal is further performed, and the area is computed. The value of the obtained area is written into a memory device as the reference value. Then, the processing is stopped for a specified time period. Then, the images are inputted again after the specified time period, and the images are averaged. Thus, the new image is obtained. The difference between the red input image and the green input image which are extracted from the obtained image is obtained. The red image is extracted, and the area of the red image is computed. The computed value of the area of the obtained red image is compared with the computed value of the area of the red image in the reference image. When, the newest computer value of the area becomes smaller than the previous computed value, the oil leakage is judged.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an oil leakage detection method suitable for use in detecting oil leakage into an oil pan below a burner gun of a boiler.

``Prior art'' Conventionally, the method of detecting oil dripping into the oil pan of a boiler, etc. is to compare a reference image taken at a certain point with an image taken at the current time to detect the spread due to oil leakage. has been adopted.

For reference, a specific example of this method is shown as a flowchart in FIG.

Problems to be Solved by the Invention" By the way, in the conventional oil leak detection method described above,
Oil leaks are detected by comparing the image (reference image) with the image, so it is applicable to items that move due to the expansion and contraction of the foiler (expands and contracts depending on the load applied), such as the oil pan under the burner gun. The drawback was that it couldn't be done.

This invention was made in view of the above-mentioned circumstances, and an object of the present invention is to provide an oil leak detection method that can reliably detect oil leaks even if the object to be detected is moving, such as the oil pan under the burner gun. The purpose is

1. A means to solve a problem.

In this invention, the area to be measured is colored, and
By imaging the area to be measured and extracting the colored color, the area is calculated at a time lag, and if the latest area calculation value is less than the previous area calculation value, it is determined that there is an oil leak. It is characterized by

"Operation" According to the method of the present invention, the area of the color colored in the measurement area is calculated with a time lag by imaging, and the area is calculated based on the latest area calculation value or the previous area calculation value. When the amount decreases, it is determined that there is an oil leak. That is, as oil drips onto the measurement area over time, the colored area of the area decreases, so it can be determined whether oil has leaked.

"Embodiments" Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the electrical configuration of an oil leak detection device to which an oil leak detection method according to an embodiment of the present invention is applied. In this figure, reference numeral 1 denotes an industrial television camera (hereinafter referred to as an ITV camera), which is set so that the oil pan below the burner gaff (the path shown) is within its field of view. Here, in the area (measurement area) where oil drips from the burner gun (path shown) of the oil pump under the burner gun,
The oil pan is colored in a different color (for example, red) from the pre-colored color (for example, black and silver). An amplifier 3 amplifies the composite color video signal supplied from the ITV camera I. 4 is the time base collector,
Composite color video signals output from the amplifier 3 at unequal intervals are corrected to equal intervals. A distributor 5 supplies the composite color video signal supplied from the time base collector 4 to a color decoder 6 and a VTR (video tape recorder) 7, respectively. The color decoder 6 separates the composite color video signal into RGB signals. Reference numeral 8 denotes an image processing device which reads RGB signals supplied from the color decoder 6 and performs predetermined image processing based on the RGB signals. Reference numeral 9 denotes a processed image monitor, which displays the processing status of the image processing device 8. IO is an encoder that restores the RGB signal to a composite color video signal, and 11 is a VTR 11 that records the decoded color video signal supplied from the encoder IO.
2 is a raw image monitor for reproducing decoded color video signals recorded in each of the VTRs 7 and +1. 13 is a controller that controls the image processing device 8, and a CPU (not shown)
(Central Processing Unit), ROM (Read Only Memory), R
It is configured with AM (random access memory), an interface circuit, and the like. In this case, a program for controlling the CPU is written in the ROM.

Next, the operation of the oil leak detection device having the above configuration will be explained with reference to the flowchart shown in FIG. Note that since the image processing device 8 described above operates according to instructions from the controller 13, it is necessary to use the controller 13 as the subject, or for convenience, the image processing device 8 will be treated as the subject in the following description.

Now, when the operator turns on the power switch (not shown) of the oil leak detection device, an image is first input. That is, a composite color video signal is output from the ITV camera 1, and after this is amplified by the amplifier 3,
It is supplied to the time base collector 4. After time base correction is performed by the time base collector 4, the signal is supplied to the color decoder 6 and the VTR 7 via the distributor 6. When a composite color video signal is supplied to the color decoder 6, this signal is separated into RGB signals.

The RGB signals are then supplied to the image processing device 8.

When the RGB signal is supplied to the image processing device 8, the device 8
A total of five images are input at 1/30 second intervals, and then the five input images are averaged to create a reference image (step Sl). Then, the created reference image is written into its own memory (path shown).

Next, the image processing device 8 performs red color extraction processing from the created reference image (step S2). That is, first, a light input image and a green input image are extracted, and then the difference between these light input images and the green input image is determined to extract red color. Next, in step S, the extracted red color is binarized using a well-known discriminant analysis method.
3) Further, noise removal processing is performed (step S4).

That is, noise is removed by performing contraction, microparticle removal, and expansion.

Next, the image processing device 8 performs area calculation after performing noise removal processing (step S5). Then, the obtained area value is written into its own memory (not shown) as a reference value. In this way, the area of the red image in the reference image is calculated.

Next, the image processing device 8 processes the image for a certain period of time (for example, 60 minutes).
The process is stopped (step S6). After 60 minutes have elapsed, image input is performed again (step S7). In the second case, a total of five images are input at 1/30 second intervals in the same manner as above, and these are further averaged to create a new image. Next, a light input image and a green input image are extracted from the image obtained by averaging processing, and then a difference between the extracted light input image and the green input image is determined, and a red image is extracted (step S8). . Next, the area of the red image is calculated by performing the same processing as in steps 62 to S5 described above (steps 89 to 511).

Then, the calculated value of the area of the obtained red image is compared with the calculated value of the area of the red image in the reference image described above (step 512). Then, an abnormality is determined based on the comparison results (step 513). In this case, oil drips onto the measured area above the oil tank under the burner gun, and the calculated area of the red image just obtained becomes smaller than the calculated area of the reference image, and the rate of decrease is determined in advance. If the value falls below the set value, it is determined that there is an abnormality. Conversely, if it is above the set value, it is determined to be normal. Here, the set value is, for example, a value obtained by applying 70 when the area of the reference image is 100. If it is determined to be normal, the process returns to step S6 and the operation is stopped for a predetermined period of time. If it is determined to be abnormal, whether it is the first time or the second time.
If it is determined that the first time is abnormal (step 514), the process returns to step S7, and this step S7
The processes from to step S13 are performed again, and a second determination is made. If it is determined that both times are abnormal, an alarm is sounded to notify the operator of the abnormality (step 514).

This makes the operator aware that an oil leak has occurred.

In this way, oil under Burnagan! (By coloring the area to be measured with a red color different from the color of the same oil pano, imaging the area to be measured and extracting only the red color, and then determining the temporal change in the area of the extracted color) ,
Detects oil leaks.

In the above embodiment, the reference area was determined from the image, but the area of the area to be measured under the burner gun (oil nozzle) may be measured in advance using a scale or the like, and this may be used as the reference area.

Further, in the above embodiments, a red image is processed, but an image of a color other than red (green or blue) may be processed.

Further, in the above embodiment, the present invention was applied to detecting oil leaks dripping into the oil pan under the burner gun, but it can also be applied to other locations where liquid leaks occur.

[Effect of the invention j As explained above, according to the oil leak detection method of the present invention, by extracting a specific color (red in the example) from the measurement area, the area is calculated at a time difference, If the latest area calculation value is less than the previous area calculation value, it is determined that there is an oil leak, so if oil of a color different from the specified color leaks, this can be detected with high accuracy. Since only a specific color is extracted, it is possible to reliably confirm the presence of a damaged area even if it moves.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing seven electrical components of an oil leak detection device to which the oil leak detection H8 method, which is an embodiment of the present invention, is applied, and Fig. 2 is a block diagram showing the operation of the device. FIG. 3 is a flow chart showing a conventional oil leak detection method. ] --I TV camera, 8 Image processing device, 13
controller.

Claims (1)

[Claims] The area to be measured is colored, and the area is calculated at a time lag by imaging the area and extracting the colored color, so that the latest area calculation value is the same as the previous area. An oil leak detection method characterized in that an oil leak is determined when the amount is less than a calculated value.