JP2019036889A - Image processing apparatus and image processing method - Google Patents

Abstract

To provide an image processing apparatus and an image processing method capable of improving visibility of a monitoring target in a high dust environment.SOLUTION: The image processing apparatus generates and outputs an integrated image by performing integration processing on a plurality of pieces of image data of a monitoring target in a high dust environment. In the integrated image, contrast components that fluctuate with time due to the influence of dust are averaged, and the visibility of the monitored object is improved.SELECTED DRAWING: Figure 6

Description

  The present disclosure relates to an image processing apparatus and an image processing method for processing an image obtained by imaging a monitoring object in a high dust environment.

  Since a fuel such as pulverized coal is used in a combustion furnace of a boiler facility, fine particles such as combustion ash contained in the pulverized coal adhere to the wall surface of the combustion furnace. In such a boiler facility, if abnormal deposits are found by periodically monitoring the inside of the combustion furnace, it is necessary to stop the facility and take measures such as removing the deposits.

  For example, the deposits in the combustion furnace are monitored by imaging a wall surface to be monitored using an imaging device. Inside the combustion furnace that is in operation, there is a high dust environment where fine particles such as pulverized coal float, and there is a powerful flame. is important. Although patent document 1 is not the technique regarding the combustion furnace of boiler equipment, the fog removal apparatus for removing a fog from a captured image is disclosed in the condition with the fog which consists of the floating particle which disperses in air | atmosphere. In Patent Document 1, image processing for removing the influence of fog is performed by calculating the illumination light component and the reflected light component using the fog density calculated from the illumination light component.

International Publication No. 2015/189874

  Unlike the atmosphere in which fog is generated as in Patent Document 1 described above, a powerful flame exists in the combustion furnace in operation. For this reason, it is difficult to use a camera in the visible light band as an imaging device for monitoring, and an infrared camera that captures an image in a wavelength band (about 4 μm) that avoids the emission wavelength of the flame is used. Since the light source in the infrared wavelength band is mainly the thermal radiation of the object, the imaging light acquired by the infrared camera includes i) random light emission from dust floating in the combustion furnace, and ii) from the wall of the combustion furnace. And iii) light emission of the monitored object. Therefore, in order to increase the visibility of the monitoring object, it is important to accurately grasp the elements of iii). Conversely, it is necessary to eliminate the effects of i) and ii).

  Moreover, since the above-mentioned patent document 1 is image processing in the case of imaging by irradiating the imaging target with irradiation light, as described above, the target that emits imaging light (infrared light) from the monitoring target itself by thermal radiation It cannot be applied as it is.

  At least one embodiment of the present invention has been made in view of the above circumstances, and an object thereof is to provide an image processing apparatus and an image processing method capable of improving the visibility of a monitoring target in a high dust environment.

(1) An image processing apparatus according to at least one embodiment of the present invention includes an imaging apparatus capable of imaging a monitoring object in a high dust environment in a predetermined integration time and the imaging apparatus in order to solve the above problem. An accumulation unit that accumulates a plurality of captured image data, and an accumulation process that reduces the contrast of luminance components that fluctuate over time by performing an accumulation process on the plurality of image data accumulated in the accumulation unit. An integration processing unit that generates an image; and an output image generation unit that generates an output image capable of displaying the contrast of the monitoring object based on the integration image.

  According to the configuration of (1) above, an integrated image is formed by performing integration processing on a plurality of pieces of image data captured by the imaging device with a short shutter time (integrated time). In such an integrated image, the contrast components that vary with time due to the influence of dust or the like floating in the combustion furnace are averaged. At that time, the light that reaches the camera from the monitored object in the combustion furnace is due to the heat radiation of the object itself, and the heat radiation and light emission of other structures and high-temperature suspended particles are reflected from the object surface. It became. Furthermore, scattering by suspended particles and radiation and emission of suspended particles are superimposed, and the brightness difference becomes extremely small. For this reason, in general, when averaging is performed by a method based on the shutter speed or integration time of the camera, a luminance difference that can be detected by the camera may not be obtained. In particular, when a video signal is output as a digital value, it is quantized and the contrast of the monitoring object cannot be obtained. On the other hand, a contrast signal cannot be obtained even in an image captured with a short shutter time due to quantization or the like. However, if the influence of airborne dust is statistically uniform regardless of location, the video signal is When quantized, a portion having a high luminance of the object has a stochastic high value, whereas a portion having a low luminance of the object has a low probability. As a result, in the integrated image using images captured with a short shutter time, the luminance change of the monitored object differs from the expected value when the luminance is quantized with respect to the luminance centered on the average luminance due to suspended dust. And the visibility of the monitoring object in the background of dust can be improved.

(2) In some embodiments, in the configuration of the above (1), the imaging device includes an imaging device having a plurality of pixels, and photoelectrically outputs light from the monitoring target imaged on the imaging device. The image data is output by converting and periodically transferring the charges of the plurality of pixels, and the image processing device averages the luminance distribution included in the integrated image for each transfer cycle of the imaging device. The image processing apparatus further includes an offset processing unit that performs offset processing by subtracting the obtained offset image from the integrated image.

  When an imaging device that photoelectrically converts imaging light on the imaging element and transfers charges for each pixel in parallel is used, the captured image includes a periodic offset due to the parallel number. According to the configuration of (2) above, the periodic offset included in the accumulated image is reduced by performing the offset process of subtracting the offset image obtained by averaging the luminance distribution for each parallel period of the imaging device from the accumulated image. Can be eliminated. As a result, the visibility of the monitoring object in the background of dust can be further improved.

(3) In some embodiments, in the configuration of (2) above, the offset processing is performed on a selected region of the integrated image.

  According to the configuration of (3) above, the offset process is limitedly performed on a selected region of the accumulated image. This reduces the calculation processing load compared to the case where the offset processing is performed on the entire integrated image, thereby contributing to the cost reduction of the arithmetic device for performing the processing and improving the image processing speed. Can do.

(4) In some embodiments, in the above configuration (2) or (3), the imaging device includes a plurality of amplifier circuits that amplify the charges transferred from the plurality of pixels, respectively.

  In an imaging apparatus having a plurality of amplifier circuits that respectively amplify charges transferred from a plurality of pixels, an offset image resulting from a characteristic difference (individual difference) between these amplifier circuits is included in the integrated image. In general, an offset for each pixel is set by an initial or pre-use adjustment using a reference photographing body, but the value is quantized to the minimum luminance unit, and a correction error equal to or less than the minimum luminance unit remains. At this time, when shooting an object whose luminance varies, the sum of the luminance value of the object and the correction error is quantized in the minimum luminance unit. However, when a large number of images are integrated, the correction error is larger than the retransmission luminance unit. Increases and decreases the visibility of the image. According to the configuration of (4) above, even when such an imaging apparatus is used, the offset image is eliminated by performing the offset process as described above, and the monitoring object in the background of dust Visibility can be improved.

(5) In some embodiments, in any one of the configurations (1) to (4), the high dust environment is a combustion furnace using pulverized coal as fuel.

  According to the configuration of the above (5), by applying it to an imaging device for monitoring a combustion furnace using pulverized coal as fuel, fine particles such as pulverized coal float, and even in an environment where a strong flame exists. The object to be monitored can be accurately recognized.

(6) In some embodiments, in any one of the configurations (1) to (5), the predetermined integration time is 40 milliseconds or less.

  In the configuration of (6) above, the predetermined integration time is desirably 1 millisecond or less.

(7) In order to solve the above problems, an image processing method according to at least one embodiment of the present invention captures an image of a monitoring object in a high dust environment in a predetermined integration time, and images in the imaging step An accumulation process for accumulating the plurality of image data, and an accumulation image in which the contrast of the luminance component that varies with time is reduced by performing an accumulation process on the plurality of image data accumulated in the accumulation process And an output image creating step for creating an output image capable of displaying the contrast of the monitoring object based on the accumulated image.

  According to the method (7), an integrated image is formed by performing an integration process on a plurality of imaged image data. In such an integrated image, a contrast component that varies temporally due to the influence of dust or the like floating in the combustion furnace having a contrast separated in a single image is averaged. As a result, in the integrated image, the luminance change centering on the average luminance is emphasized, and the visibility of the monitoring object in the background of dust can be improved. At this time, when the shutter time is long and time integration is performed in a single image, the visibility of the monitoring object in the background of the dust cannot be improved due to the influence of luminance quantization for each pixel.

(8) In some embodiments, in the method of (7) above, the imaging step includes an imaging device having a plurality of pixels, and photoelectrically outputs light from the monitoring object imaged on the imaging device. The image processing method uses an imaging device that outputs the image data by converting and periodically transferring charges of the plurality of pixels, and the image processing method uses the luminance distribution included in the integrated image for each transfer cycle of the imaging device. Further, an offset processing step of performing offset processing by subtracting an offset image obtained by averaging from the integrated image is further provided.

  According to the method (8), the offset included in the accumulated image can be eliminated by performing the offset process on the accumulated image. As a result, the visibility of the monitoring object in the background of dust can be further improved.

(9) In some embodiments, in the method of (7) or (8), the predetermined integration time is 40 milliseconds or less.

  In the method (9), preferably, the predetermined integration time is 1 millisecond or less.

  According to at least one embodiment of the present invention, it is possible to provide an image processing apparatus and an image processing method capable of improving the visibility of a monitoring target in a high dust environment.

It is a lineblock diagram showing boiler equipment roughly. It is a measurement example which shows the wavelength distribution contained in the light emission from the combustion furnace of FIG. It is the schematic which shows the structure of the imaging device of FIG. It is a schematic diagram for demonstrating a mode that an electric charge is transferred from each pixel of FIG. 1 is a configuration diagram illustrating an internal configuration of an image processing apparatus according to at least one embodiment of the present invention. It is a flowchart which shows the image processing method implemented by the image processing apparatus of FIG. 5 for every process. It is an example of an image in each process of Drawing 5. It is a flowchart which shows the subroutine of the offset process implemented by step S12 of FIG. It is explanatory drawing of FIG. It is an example of the monitoring object area | region selected by the integration image.

Hereinafter, some embodiments of the present invention will be described with reference to the accompanying drawings. However, the dimensions, materials, shapes, relative arrangements, etc. of the components described in the embodiments or shown in the drawings are not intended to limit the scope of the present invention, but are merely illustrative examples. Absent.
For example, expressions expressing relative or absolute arrangements such as “in a certain direction”, “along a certain direction”, “parallel”, “orthogonal”, “center”, “concentric” or “coaxial” are strictly In addition to such an arrangement, it is also possible to represent a state of relative displacement with an angle or a distance such that tolerance or the same function can be obtained.
For example, an expression indicating that things such as “identical”, “equal”, and “homogeneous” are in an equal state not only represents an exactly equal state, but also has a tolerance or a difference that can provide the same function. It also represents the existing state.
For example, expressions representing shapes such as quadrangular shapes and cylindrical shapes represent not only geometrically strict shapes such as quadrangular shapes and cylindrical shapes, but also irregularities and chamfers as long as the same effects can be obtained. A shape including a part or the like is also expressed.
On the other hand, the expressions “comprising”, “comprising”, “comprising”, “including”, or “having” one constituent element are not exclusive expressions for excluding the existence of the other constituent elements.

First, a description will be given of a boiler facility 1 serving as a monitoring target of an image processing apparatus 100 according to at least one embodiment of the present invention. FIG. 1 is a configuration diagram schematically showing a boiler facility 1.
The boiler facility 1 includes a combustion furnace (furnace) 2 and a combustion device 4 provided in the combustion furnace 2. The combustion device 4 can supply a solid powder fuel and an oxygen-containing gas to the combustion furnace 2, and the solid powder fuel burns inside the combustion furnace 2 to generate a high-temperature gas (combustion gas). The hot gas heats water as a heat medium via a heat exchanger such as a economizer, superheater, and reheater (not shown), and uses the steam obtained thereby, for example, turbine power generation (not shown). The machine can be activated.

  The solid powder fuel is a fine powder fuel produced by pulverizing a solid raw material, for example, a simple substance or a mixture of coal, oil coke, solid biomass and the like. In the present embodiment, the solid powder fuel is pulverized coal, and the pulverized coal is obtained when the mill 6 provided in the boiler facility 1 pulverizes the coal.

The combustion apparatus 4 includes at least one burner 8 that can be attached to the combustion furnace 2 and a wind box 10 that can be attached to the combustion furnace 2 so as to surround the burner 8. The pulverized coal is carried by the carrier gas sent from the blower 12 and supplied to a fuel supply nozzle (not shown) of the burner 8. On the other hand, oxygen-containing gas is sent from the blower 13 to the wind box 10. The carrier gas and the oxygen-containing gas are, for example, air.
A part of the carrier gas and the oxygen-containing gas can be heated to an appropriate temperature by the heater 14. The heater 14 may be incorporated in the boiler facility 1.

  In such boiler equipment 1, a monitoring window 16 for monitoring the wall surface of the combustion furnace 2 is provided. The monitoring window 16 is a transparent window on the side facing the wall surface to be monitored. An imaging device 20 is installed outside the combustion furnace 2 and is configured to be monitored by imaging the wall surface of the combustion furnace 2 through the monitoring window 16. Since such an imaging device 20 is installed outside the high-temperature combustion furnace 2, the inside of the combustion furnace 2 can be monitored without being exposed to a harsh environment.

  Moreover, in the combustion furnace 2 in operation, a powerful flame exists by burning pulverized coal. The imaging device 20 performs imaging in a wavelength region that avoids the emission wavelength of the flame in order to eliminate the influence of such emission of the flame. Here, FIG. 2 is a measurement example showing the wavelength distribution included in the light emission from the combustion furnace 2 of FIG. As shown in FIG. 2, the emission wavelength of the flame is small in the infrared wavelength band of about 4 μm, and the imaging device 20 performs imaging in the wavelength band.

  FIG. 3 is a schematic diagram showing the configuration of the imaging device 20 of FIG. The imaging device 20 includes an imaging element 24 having a plurality of pixels 22. Specifically, the image sensor 24 is a CMOS image sensor, for example, and is a matrix type image sensor having a plurality of pixels 22. In the plurality of pixels 22, the light imaged on the image sensor 24 from the monitoring target is photoelectrically converted. The charges converted in each pixel 22 are sequentially transferred by the vertical scanning circuit 26 and the horizontal transfer scanning circuit 28. The transferred charge is amplified by the amplifier circuit 30 and then output to the outside as image data.

  The transfer of charges from each pixel 22 is periodically performed by the vertical scanning circuit 26 and the horizontal transfer scanning circuit 28. FIG. 4 is a schematic diagram for explaining how charges are transferred from each pixel 22 in FIG. Here, as shown in FIG. 4A, a matrix type including (m × n) pixels 22 is assumed as the image sensor 24. The shutter open time or the charge accumulation time of the element was set to a short time of 40 milliseconds or less, and in this example, 30 microseconds.

When the vertical scanning circuit 26 selects a specific row, the horizontal transfer scanning circuit 28 simultaneously transfers charges from the adjacent N pixels 22 in the selected row, and repeats until the transfer in the row is completed. That is, as shown in FIG. 4B, when the m1th row is selected, the first transfer includes the pixels 22 ₁ , 22 ₂ ,. At the same time the charge from the _N is transferred, followed by (N + 1) ~2N th pixel _{22 N + 1, 22 N +} 2, at the same time the charge from · · · _{22 2N} are transferred. This is repeated until the transfer of charges from the pixel 22 in the last column (nth column) in the m1th row is completed.

When the transfer of charges in the m1th row is completed in this way, the vertical scanning circuit 26 selects an adjacent row, and the horizontal transfer scanning circuit 28 selects N adjacent pixels in the selected row as described above. At the same time, charge transfer is repeated. That is, as shown in FIG. 4C, charges are transferred simultaneously from m rows 22 ₁ , 22 ₂ ,... 22 _N 22 for which transfer has been completed, and then (N + 1) to 2N-th pixels 22 are transferred. Charges are transferred from _{N + 1} , 22 _{N + 2} ,... 22 _2N . This is repeated until the transfer of charges from the pixel 22 in the last column (n-th column) in the (m1 + 1) -th row is completed.
In this manner, periodic transfer is sequentially performed over the entire (m × n) pixels included in the image sensor 24.

  When the inside of the combustion furnace 2 is imaged using such an imaging device 20, the visibility of the observation object changes according to the concentration of dust floating in the combustion furnace 2. Further, since the dust concentration fluctuates with time inside the combustion furnace 2 (because the dust concentration distribution changes), the visibility of the imaging device 20 varies with time. That is, the dust concentration is reduced at a certain moment and the visibility is improved, but the dust concentration is increased and the visibility is lowered at another moment. For this reason, it is appropriate to set the image integration time to 25 milliseconds or less, more suitably 1 millisecond or less.

  Thus, since the captured image of the imaging device 20 fluctuates according to the dust concentration of the combustion furnace 2, it is difficult to recognize the monitoring target well with a single captured image. Accordingly, the image data acquired by the imaging device 20 is sent to an image processing device 100 described below, whereby predetermined image processing is performed.

  5 is a configuration diagram of the image processing apparatus 100 according to at least one embodiment of the present invention, and FIG. 6 is a flowchart showing an image processing method performed by the image processing apparatus 100 of FIG. These are examples of images in each step of FIG.

The image processing apparatus 100 is composed of an electronic arithmetic device such as a computer, for example, and a program for performing image processing to be described later is installed. Such an image processing apparatus 100 includes a storage unit 102 that stores various data, an integration processing unit 104 that performs integration processing on a plurality of image data stored in the storage unit 102, and an offset that performs offset processing. A processing unit 106 and an output image creation unit 108 that performs an image adjustment process to create an output image are provided.
In the present embodiment, the storage unit 102 includes a storage device such as a hard disk or a memory, and a plurality of storage units 102 are prepared according to the type of data to be stored. Specifically, a first accumulation unit 102a for accumulating a plurality of image data captured by the imaging device 20, and a second accumulation unit 102b for accumulating the accumulated image created by the accumulation processing unit 104. And a third accumulation unit 102c for accumulating the image data subjected to the offset process, and a fourth accumulation unit 102d for accumulating the output image.

  First, the image processing apparatus 100 acquires a plurality of image data of a monitoring target (a wall surface of the combustion furnace 2) by continuously capturing images in time series with the imaging apparatus 20 (step S10). The imaging by the imaging device 20 is performed in the infrared wavelength band as described above with reference to FIG.

FIG. 7A is an example of image data stored in the first storage unit 102a. Thus, in the image data acquired from the imaging device 20, the state of the wall surface of the combustion furnace 2 to be monitored is unclear due to the influence of dust, which is insufficient for monitoring.
Note that an adaptive histogram averaging process that averages the contrast for each minute area (for example, a 9 × 9 pixel area) may be performed on such image data. Thereby, visibility can be improved when the brightness distribution in an image is large.

  As described above with reference to FIGS. 2 and 3, the imaging device 20 outputs image data by performing periodic transfer from the imaging device 24. A plurality of image data output from the imaging device 20 is readable and stored in the first storage unit 102a.

  The plurality of image data stored in the storage unit 102 is read by the integration processing unit 104, and integration processing is performed (step S11). The integration process is performed by adding the luminance values of the corresponding pixels 22 to the plurality of image data read from the first accumulation unit 102a. In such integration processing, contrast is emphasized for a monitoring object (a wall surface of the combustion furnace 2) that remains unchanged in time, while a fluctuation component that fluctuates in time is canceled (averaged) around an average value. And the contrast is reduced. Therefore, in the integrated image created in the integration process, the influence of the dust concentration that fluctuates inside the combustion furnace 2 is eliminated, and the visibility of the monitoring object is improved. The accumulated image created by the accumulation processing unit 104 in this way is accumulated in the second accumulation unit 102b so as to be readable.

  FIG. 7B is an example of an accumulated image formed by the accumulation processing unit 104. In the integrated image, the unevenness of the wall surface of the combustion furnace 2 that could not be recognized due to the influence of dust in the single image of FIG. 7A is displayed so as to be raised, and the visibility is improved. In such an integrated image, if the wall surface can be seen for a short time in a plurality of image data, the wall surface (wall surface nintrust or clinker that varies over a long time) when the influence of other fluctuation components is eliminated. Is effective because it is highlighted.

  In the example of FIG. 7B, the visibility of the wall surface of the combustion furnace 2 to be monitored is improved, but this image includes a striped pattern composed of a plurality of lines along the vertical direction. This is because a minute difference in offset for each pixel 22 that has been corrected to one gradation or less in the image pickup device 24 included in the image pickup device 20 is emphasized by integrating the image data. According to the earnest study of the present inventor, such a striped pattern is caused in particular by a characteristic difference (individual difference) between a plurality of amplifiers included in the amplifier circuit 30 that amplifies the image signal, and this is a subsequent offset process. It was found that it can be suppressed satisfactorily.

  The accumulated image accumulated in the second accumulation unit 102b is read by the offset processing unit 106, and an offset process is performed (step S12). The offset processing is performed based on an offset image obtained by averaging the pixels included in the integrated image for each transfer cycle of the image sensor 24 of the imaging device 20.

  Here, the offset process in step S12 of FIG. 6 will be described in detail with reference to FIG. FIG. 8 is a flowchart showing a subroutine of offset processing executed in step S12 of FIG. 6, and FIG. 9 is an explanatory diagram of FIG.

  First, the offset processing unit 106 acquires the accumulated image stored in the second storage unit 102b (step S20). In the left diagram of FIG. 9A, an example of the accumulated image acquired by the offset processing unit 106 is shown on the XY plane. Here, the X axis is set parallel to the horizontal scanning direction of FIG. 2, and the Y axis is set parallel to the vertical scanning direction of FIG.

Subsequently, as shown in the right diagram of FIG. 9A, a partial region of the integrated image acquired by the offset processing unit 106 is selected (step S21), and the luminance value of each pixel in the region is specified. Thus, a three-dimensional luminance map shown in FIG. 9B is created (step S22). Here, the three-dimensional luminance map is a three-dimensional map with the luminance value of each pixel 22 in the selected partial region (XY plane) as the Z axis.
Note that the partial region in step S21 may be selected as a region that is at least larger than the number N of charges that can be transferred simultaneously in the imaging device 20. In the present embodiment, the following description will be made assuming that an area in which (m × n) pixels 22 are arranged in a matrix is selected.

  Then, the offset processing unit 106 obtains the average value of the luminance values for the pixels 22 arranged along the Y axis in the three-dimensional luminance map of FIG. 9B (aggregates along the Y axis), and FIG. The characteristic curve shown in (c) is calculated (step S23). For example, when a total of (m × n) pixels 22 are arranged in a matrix, an average value of luminance values is obtained for m pixels 22 arranged in each column, and the average value is represented by the vertical axis and the X axis. The characteristic curve is calculated so that the coordinate of is the horizontal axis.

  As shown in FIG. 9C, the horizontal axis of the characteristic curve corresponds to the number of pixels along the horizontal direction. As described above, since the imaging device 20 transfers N charges simultaneously in each row, the characteristic curve is divided into a plurality of segments S1, S2,... S24). Then, the integrated characteristic curve shown in FIG. 9D is calculated by superimposing the plurality of divided segments S1, S2,... (Step S25). Such an integrated characteristic curve reflects the characteristics of each channel of the imaging device 20.

  Subsequently, a linear approximate straight line is obtained for the integrated characteristic curve calculated in step S25 (step S26). The integrated approximate straight line is obtained by specifying constants a and b by fitting Z = aX + b to the integrated characteristic curve. Subsequently, the correction value of the offset correction is set so that the slope a of the linear approximation line obtained in step S26 becomes zero (that is, the linear approximation line is parallel to the X axis as shown in FIG. 9E). Determine (step S27). Subsequently, a periodic averaged image is created by developing the correction value determined in step S27 in a range corresponding to the entire integrated image in step S20 (step S28). FIG. 9F is an example of the periodic averaged image created in step S28.

  Then, an offset process is performed by subtracting the periodic averaged image created in step S28 from the integrated image in step S20 (step S29). Thereby, the noise of the striped pattern displayed in FIG. 7B is appropriately removed, and the wall surface of the combustion furnace 2 to be monitored is displayed more clearly.

  In step S27, the correction value may be determined so that the linear approximation line approaches a constant luminance. Ideally, it is preferable to correct the linear approximation line so that it has a constant luminance. However, in this embodiment, the offset correction is performed so that the slope of the linear approximation line becomes zero in consideration of the balance with the calculation burden. Has been done.

  The accumulated image subjected to the offset process in this way is stored in the third storage unit 102c so as to be readable. Then, the output image creation unit 108 reads the accumulated image accumulated in the third accumulation unit 102c and appropriately adjusts the output image so as to conform to a display format in a display device (not shown) such as a display. Is created (step S13). The output image on which the adjustment process has been performed is stored in the fourth storage unit 102d and displayed on the display device 110 as appropriate (step S14).

  Note that the offset process in step S12 may be performed on a selected part of the integrated image. For example, as illustrated in FIG. 10, the monitoring target area 40 may be selected from the integrated image, and the offset process may be performed on the monitoring target area 40 in a limited manner. In this case, the calculation processing load is reduced compared with the case where the offset processing is performed on the entire integrated image, which contributes to the cost reduction of the arithmetic device for performing the processing and improves the image processing speed. Can do.

As described above, according to the present embodiment, an integration image is formed by performing integration processing on a plurality of image data captured by the imaging device 20. In such an integrated image, the contrast components that vary with time due to the influence of dust or the like floating in the combustion furnace are averaged. As a result, in the integrated image, the luminance change centering on the average luminance is emphasized, and the visibility of the monitoring object in the background of dust can be improved.
Furthermore, by performing the offset process on the integrated image, the offset due to the characteristics of the imaging device 20 can be eliminated, and the visibility of the monitoring object in the background of dust can be further improved.
In this way, it is possible to provide an image processing apparatus and an image processing method capable of improving the visibility of a monitoring target in a high dust environment.

  The present invention is applicable to an image processing apparatus and an image processing method for processing an image obtained by imaging a monitoring object in a high dust environment.

DESCRIPTION OF SYMBOLS 1 Boiler equipment 2 Combustion furnace 4 Combustion device 6 Mill 8 Burner 10 Wind box 12 Blower 14 Heater 16 Monitoring window 20 Imaging device 22 Pixel 24 Imaging element 26 Vertical scanning circuit 28 Horizontal transfer scanning circuit 30 Amplifier circuit 40 Monitoring object area 100 Image Processing device 102 Storage unit 102a First storage unit 102b Second storage unit 102c Third storage unit 102d Fourth storage unit 104 Integration processing unit 106 Offset processing unit 108 Output image creation unit

Claims (9)

An imaging device capable of imaging a monitoring object in a high dust environment in a predetermined integration time;
An accumulation unit that accumulates a plurality of image data captured by the imaging device;
An integration processing unit that creates an integrated image with reduced contrast of luminance components that vary with time by performing integration processing on the plurality of image data stored in the storage unit;
Based on the accumulated image, an output image creating unit that creates an output image capable of displaying the contrast of the monitoring object;
An image processing apparatus comprising: The imaging device includes an imaging device having a plurality of pixels, photoelectrically converts light from the monitoring target imaged on the imaging device, and periodically transfers charges of the plurality of pixels. Outputting the image data;
The image processing apparatus further includes an offset processing unit that performs an offset process by subtracting an offset image obtained by averaging the luminance distribution included in the accumulated image for each transfer period of the imaging device from the accumulated image. The image processing apparatus according to claim 1.   The image processing apparatus according to claim 2, wherein the offset process is performed on a selected part of the integrated image.   The image processing apparatus according to claim 2, wherein the imaging apparatus includes a plurality of amplification circuits that respectively amplify charges transferred from the plurality of pixels.   The image processing apparatus according to claim 1, wherein the high dust environment is a combustion furnace using pulverized coal as a fuel.   The image processing apparatus according to claim 1, wherein the predetermined integration time is 40 milliseconds or less. An imaging step of imaging a monitoring object in a high dust environment in a predetermined integration time;
An accumulation step of accumulating a plurality of image data imaged in the imaging step;
An integration process step of creating an integrated image with reduced contrast of luminance components that vary with time by performing an integration process on the plurality of image data stored in the storage step;
An output image creating step for creating an output image capable of displaying the contrast of the monitoring object based on the accumulated image;
An image processing method comprising: In the imaging step, an imaging device having a plurality of pixels is included, photoelectrically converting light from the monitoring object imaged on the imaging device, and periodically transferring charges of the plurality of pixels. Using an imaging device that outputs the image data,
The image processing method further includes an offset processing step of performing offset processing by subtracting an offset image obtained by averaging the luminance distribution included in the integrated image for each transfer period of the imaging device from the integrated image. The image processing method according to claim 7.   The image processing method according to claim 7 or 8, wherein the predetermined integration time is 40 milliseconds or less.