JPH08194805A - Picture processor and picture processing system - Google Patents

Abstract

PURPOSE: To shorten retrieving processing for the accident occurrence place and time of a power failure accident due to a comparatively small accident current by detecting a wide accident current area by plural cameras using neutral density(ND) filters having respectively different transmittance values. CONSTITUTION: This picture processor is provided with a 1st camera 2a provided with a 1st ND filter 3a and a 2nd camera 2b provided with a 2nd ND filter 3b having higher transmittance value than the 1st ND filter 3a. When an accident current is small like a grounding accident, the signal level of an output picture from the 1st camera 2a through the 1st ND filter 3a is low and its detection precision is extremely low. However a video output from the 2nd camera 2b through the 2nd ND filter 3b becomes fine picture data, so that high detection precision can be obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to improvement of the detection range performance of a short-circuit / ground fault monitoring image processing device in high-voltage power transmission / distribution electric equipment such as a substation.

[0002]

[Prior art]

Conventional example 1. FIG. 24 is a block diagram of a conventional short-circuit accident monitoring system for electric equipment in a substation or the like, which is a relay point of power supply, due to invasion of small animals. In the camera section, this camera section 2 is a neutral density filter (hereinafter referred to as N
D filter) 3, lens 4, microphone 5, camera body 6,
It is composed of a camera turntable 7. An image processing device 12 is connected to the camera unit 2 via a camera control signal line 9 and a video signal transmission line 10, and the image processing device 12 is composed of an image processing unit 13 and an image memory 14. Reference numeral 8 is a monitor device connected to the image processing device 12, and 15 is an input operation unit connected to the image processing device 12 via an input operation control signal line 16.

Further, FIG. 25 shows a model of a graph of brightness of the accident occurrence unit 1 versus sensor output. In the figure, the vertical axis is the sensor output, the horizontal axis is the brightness, the area hatched with thin lines is the brightness area in the normal state, and the area hatched with the dotted lines is the brightness area in the event of an accidental discharge. The area sandwiched between the two areas is a dead zone, and the width of this dead zone depends on the magnitude of the current to be monitored as a fault current. That is, if the lower limit value of the fault current is close to the normal region, there may be no dead zone at all, and if the lower limit value is large, the width of the dead zone becomes large. Reference numeral 31 represents a sensitivity curve of the camera 2 when the ND filter 3 is not used, and 32 represents a sensitivity curve of the camera 2 when the ND filter 3 is used.

Next, the operation will be described. In FIG. 24, when the accident occurrence unit 1 is not discharged due to an accident, the brightness input to the camera 2 is the normal area in FIG. Here, the ND filter 3 is connected to the camera 2.
Since the camera is attached, the sensitivity of the camera system as a whole becomes a sensitivity curve 32, and the sensor output is zero, that is, the screen of the monitor device 8 shows nothing. When the camera 2 does not have the ND filter 3, the sensitivity of the entire camera system is the sensitivity curve 31, so the sensor output becomes an image according to the brightness of the target image, and the image appears on the monitor device 8. When the accident occurrence part 1 is discharged due to an accident, especially when the accident current value is large like a short-circuit accident, the brightness input to the camera 2 is
It is the accident area in FIG. 25, and even if the ND filter 3 is attached, it is within the range of the sensitivity curve 32, and the place where the discharge phenomenon occurs becomes the brightest on the screen of the monitor device 8, for example, and is shown in FIG. The image that looks like appears. The image memory 14 normally displays a continuous screen in frame units, or
It is configured such that a plurality of screens can be stored in field units, and when an abnormal state does not occur, the oldest screen among the plurality of screens is pushed out and the latest screen is stored. Here, when a power failure occurs due to an accident, a power failure signal (not shown) indicating that the power failure has occurred is input to the image processing unit 13. If a power failure does not occur even if an accident occurs, a power failure signal is not generated, so the system determines that no accident has occurred and continues normal monitoring operation. That is, the video capturing sequence is repeatedly executed. The image memory 14 is controlled by the image processing unit 13 so that the designated number of screens is recorded in the image memory 14 from the time when the power failure signal is input, and thereafter the recording in the image memory 14 is stopped.

Further, when the accident current is small as in the case of a ground fault, the brightness input to the camera unit 2 is near the lower limit value of the accident area near the dead zone area in FIG. When mounted, the sensor output is small and it is difficult for a clear image to appear on the screen of the monitor device 8.

Conventional example 2. Further, Japanese Patent Laid-Open No. 64-5185 includes a plurality of cameras with filters having different transmissivities for image input, and is sensitive to light in at least two regions of ultraviolet region, visible region and infrared region. An image pickup lens that transmits light in the same region as the image pickup element having
An image segmentation camera device with a filter is described.

Conventional example 3. Furthermore, JP-A-4-12451
Japanese Unexamined Patent Publication 7 discloses a fire detector composed of a plurality of photodetectors and controllers which have detection sensitivities for different wavelength regions of light and are configured so that light is commonly incident. ing.

[0008]

Since the conventional image processing system for detecting a short-circuit accident is constructed as described above,
Only a fault with a relatively large current such as a short-circuit fault can be detected, and a fault with a relatively small current such as a ground fault cannot be detected.

The present invention has been made in order to solve the above problems, and is provided with a plurality of cameras using ND filters having different transmissivities and sharing the capacity of an image memory reduces the cost. It is intended to improve the accuracy of accident detection while suppressing the rise.

[0010]

According to another aspect of the present invention, there is provided an image processing apparatus, comprising a plurality of cameras having ND filters having different transmittances, for monitoring a short circuit / ground fault accident of power equipment, and recording image signals of the cameras. And image analysis processing means for analyzing the image signal stored in the image storage means.

According to a second aspect of the present invention, in the image processing apparatus according to the first aspect, a first camera having a low transparency and a second camera having a high transparency are provided. .

According to a third aspect of the present invention, in the image processing apparatus according to the first aspect, the image storage means records the image signals before and after the power failure signal is generated.

According to a fourth aspect of the present invention, in the image processing apparatus according to the first aspect, the image analysis processing means specifies a frame in which an image in which a short circuit / ground fault has occurred is recorded, and which part of the specified frame is specified. Is characterized by searching for the brightest.

An image processing apparatus according to a fifth aspect is the image processing apparatus according to the first aspect, further comprising a monitor device for displaying a result retrieved by the image analysis means.

An image processing system according to a sixth aspect of the present invention is an N system in which the number of frames and the transparency set after the power failure signal is generated are different.
Image recording of image signals of a plurality of cameras having a D filter is continued, the frame in which the image in which the accident has occurred is recorded is identified from the captured image record, and which part of the identified frame is brightest is searched. , The search result is displayed on the monitor screen.

The image processing system according to claim 7 is the image processing system according to claim 6.
In the above description, the area on the monitor screen is divided and the process of searching for the brightest block is repeated to perform the place search process.

The image processing system according to claim 8 is the image processing system according to claim 7.
In the described one, the process of searching the brightest block by dividing the 64 blocks in the area on the monitor screen into four
It repeats once.

The image processing system according to claim 9 is the image processing system according to claim 7.
In the described ones, candidate blocks in the vicinity other than the brightest block are searched and displayed.

An image processing system according to a tenth aspect of the present invention is the image processing system according to the ninth aspect, wherein the threshold value 1 for searching the brightest fixed block next to the brightest fixed block is set as a parameter, and the brightness of the fixed block is set. The threshold value obtained by multiplying the threshold value with the brightness of the blocks above and below, and to the left and right of the fixed block is determined as a candidate block.

An image processing system according to an eleventh aspect is the image processing system according to the sixth aspect, wherein the video data from the first camera is assigned from the first scanning line to a predetermined scanning line on the frame memory, and the image data from the second camera is assigned. Video data is assigned to the remaining scanning lines and the same timing frame is recorded in the same frame memory.

An image processing system according to a twelfth aspect of the present invention is the image processing system according to the sixth aspect, wherein the frame search processing is performed so that the time is traced back from the frame image data at the time when the power failure signal is generated.

According to a thirteenth aspect of the present invention, in the image processing system according to the sixth aspect, the frame search processing is performed by thinning out the data on the frame memory in the horizontal scanning line direction.

According to a fourteenth aspect of the present invention, in the image processing system according to the sixth aspect, the frame search processing is performed by thinning out the data on the frame memory in the pixel direction.

According to a fifteenth aspect of the present invention, in the image processing system according to the sixth aspect, the frame search processing is performed by thinning out the data on the frame memory.

According to a sixteenth aspect of the present invention, in the image processing system according to the seventh aspect, the image processing system has a plurality of cameras each having an ND filter having a transmittance that is a power ratio, and a preset optimum image brightness. Is compared with the brightness of the area to which one of the cameras is assigned, and the camera to be compared next is determined based on the result.

[0026]

According to the image processing apparatus of the present invention, N having different transparency is used.
A camera equipped with a D filter detects a wide range of fault current regions.

An image processing apparatus according to a second aspect of the present invention detects a wide range of fault current regions by using a first camera having a low transparency and a second camera having a high transparency.

The image processing apparatus of the third aspect records the image signals before and after the power failure signal is generated.

An image processing apparatus according to a fourth aspect of the present invention identifies a frame in which an accident has occurred and searches which part of the identified frame is the brightest.

In the image processing apparatus according to the fifth aspect, the monitor displays the result retrieved by the image analysis means.

According to another aspect of the image processing system of the present invention, a frame in which an accident has occurred is identified from image records of a plurality of cameras captured after a power failure signal is generated, and which part of the identified frame is brightest. Search and display the results on the monitor screen.

In the image processing system according to the seventh aspect, the area on the monitor screen is divided and the processing for searching the brightest block is repeated to perform the location search processing.

The image processing system of claim 8 repeats three times the process of dividing the 64 blocks of the area on the monitor screen into four and searching for the brightest block.

The image processing system according to claim 9 searches and displays candidate blocks in the vicinity other than the brightest block.

An image processing system according to a tenth aspect of the present invention compares the threshold value 1 for searching the brightest block next to the brightest fixed block with the brightness of the upper, lower, left and right blocks of the fixed block to obtain the threshold value 1 The brighter block is the candidate block.

According to the image processing system of the eleventh aspect, the same timing frame is recorded in the same frame memory.

In the image processing system according to the twelfth aspect, the frame search processing is performed to trace back the time from the frame image data at the time when the power failure signal is generated.

In the image processing system according to the thirteenth aspect, the frame search processing is performed by thinning out the data on the frame memory in the horizontal scanning line direction.

In the image processing system according to the fourteenth aspect, the frame search processing is performed by thinning out the data in the frame memory in the pixel direction.

The image processing system according to the fifteenth aspect performs the frame search processing by thinning out the data in the frame memory.

According to a sixteenth aspect of the present invention, in the image processing system, the average value of the brightness during the optimum image is compared with the brightness of the area to which one camera is assigned, and the ND filter in which the transmittance is a power ratio is used. The next camera to be compared is determined.

[0042]

【Example】

Example 1. Embodiment 1 of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of an image processing system composed of two sets of cameras. In the figure, 1 is an accident occurrence part, 2a is a first camera part, a first ND filter 3a, a first lens 4a, and a first lens 4a. 1 microphone 5a, first camera body 6a,
It is composed of a first camera turntable 7a. Reference numeral 2b denotes a second camera unit, which, like the first camera unit 2a, has a second N
D filter 3b, second lens 4b, second microphone 5
b, the second camera body 6b, and the second camera turntable 7b. An image processing device 12 includes first and second image processing devices.
Camera control signal lines 9a and 9b to the camera units 2a and 2b, respectively.
b, connected via video signal transmission lines 10a, 10b,
The image processing device 12 includes an image processing unit 13 and an image memory 14. Reference numeral 8 is a monitor device connected to the image processing device 12, and 15 is an input operation unit connected to the image processing device 12 via an input operation control signal line 16.

Next, the operation will be described. FIG. 2 shows a flowchart of the entire processing of the image processing system in this embodiment. Figure 2 does not show the case where no accident has occurred. Only the processing after the power failure signal (not shown) is displayed. As a rough processing flow, image recording is continued for a set number of frames (6 frames in the present embodiment) after the power failure signal is generated. That is, 10 frames before the power failure signal is generated and 6 frames after the power failure signal are captured in the image memory (step 20). Subsequently, the process moves to the analysis process of the image data in the captured image memory. First, as the first step of the analysis process, a frame search process is performed to identify the frame in which the video in which the short circuit accident has occurred is recorded (step 21). Next, the image data of the frame specified in the pre-processing is specified to specify the location of the accident, and a search process is performed to find out which part of the frame is brightest (step 22).
Then, the retrieved result is displayed on the monitor screen as a processing result, and a series of processing is ended (step 23).

A supplementary explanation of each of the above processes will be given below. First, image capturing is processed by the procedure shown in FIG. That is, when a power failure occurs due to discharge due to an accident, a power failure signal is input to the image processing device 12, and from that time point, a specified number of frames (here, description is made of 6 frames) of video signals are continuously recorded, After that, the video recording operation is stopped. That is, the video data before the occurrence of the power failure signal is recorded in the image memory for 10 frames, and the video data after the occurrence of the power failure signal is recorded in the image memory for 6 frames. In FIG. 1, the second ND filter 3b is the first ND filter 3a.
It is assumed that the transmittance is higher and the relationship between the sensor output and the brightness at that time is as shown in FIG. 7. In the normal operation state, that is, when the accident occurrence part 1 is not discharged due to the accident, the first and second camera parts 2a,
The brightness input to 2b is in the normal region in FIG. 7, and even when the second ND filter 3b having high transparency is attached, the sensor output is zero from the sensitivity curve 42, that is, the monitor device 8 The screen becomes black,
It means that there is nothing. Then, the accident occurrence part 1
If a discharge occurs due to an accident, like a ground fault,
When the fault current is relatively small, the sensitivity curve 42 is generated by the second ND filter 3b, and particularly, as in the case of a short circuit fault,
When the accident current value is large, the sensitivity curve 43 is set by the first ND filter 3a, and the camera 2
The images captured in a will appear as images as shown in FIG. 6 on the screen of the monitor device 8.

Here, when the fault current is small like a ground fault, the output image from the first camera 2a through the first ND filter 3a is shown in FIG. Since the output image from the second camera 2b that has passed through the filter 3b becomes image data equivalent to the image captured by the first camera in FIG. 6, in most cases, the signal level is low in the first camera output image. , If this is left as it is, the detection accuracy will be extremely low. However, the video output from the second camera 2b through the second ND filter 3b becomes good image data, and as a result, high detection accuracy can be obtained.

On the contrary, when the accident current is large like a short circuit accident, the output image from the first camera 2a through the first ND filter 3a is equivalent to the image captured by the first camera in FIG. The output image from the second camera 2b, which is the same image as the data, passed through the second ND filter 3b is
As shown in FIG. 9, in the second camera output image, the signal level is high over the entire screen, and as a result, the dynamic range is reduced, and the detection accuracy is extremely low as it is. However, the image output from the first camera 2a through the first ND filter 3a becomes good image data, and as a result, high detection accuracy can be obtained.

In the subsequent processing of image data, the image memory 14 is usually constructed so that a plurality of screens can be distributed and stored on a frame-by-frame basis or on a field-by-field basis as in the conventional example, so that no abnormal state occurs. In this case, the oldest screen among the plurality of screens is pushed out and the latest screen is stored. Here, when a power failure signal (not shown) indicating that a power failure has occurred due to an accident is input to the image processing unit 13, the designated number of screens is recorded in the image memory 14 from the time when the signal is input, and thereafter. The image processing unit 13 controls the image memory 14 so that the recording to the image memory 14 is stopped.

Next, a frame search method will be described with reference to the frame search flowchart of FIG. In this embodiment, as the frame search method, the brightest frame is defined as the frame in which the image at the time of the accident is recorded, and the processing is performed from the recorded image data of each frame. First, the total brightness of the video data of the first frame is calculated, and the average value of the brightness is obtained (steps 40 to 42). Noise can be expected to be removed by averaging. Next, the average value of the brightness of the second frame is similarly obtained, and the average value of the brightness of the first frame is compared with the size (step 43). At this time, in order to remove an error due to a slight change in brightness, the determination is made based on whether or not there is a difference over a certain change width (denoted by α in this embodiment) (whether or not there is a decrease). . The same process is repeated until a change occurs. If all 1
If no change is obtained even after performing the processing for 6 frames, it is determined that no accident has occurred, the result is displayed on the monitor screen, and the processing ends. When a change occurs in the Nth frame, the previous frame, that is, N
It is determined that the -1st frame is the maximum brightness frame, the result is recorded as a T frame, and the process proceeds to the next location search process.

Next, the place search method will be described with reference to FIGS. In this embodiment, the area on the screen is divided into 8 parts in the vertical direction and 8 parts in the horizontal direction, and 64 blocks are divided into 4 parts, and the same processing as the first time is performed. In this example,
An example in which the third block is brightest is shown. The detected block number is registered as the Y coordinate, and the process proceeds to the third processing. Next, similarly, the third processing is performed, and the detected block number is registered as the Z coordinate. 6 in total
When the finally detected block among the four blocks is represented by the coordinates (X, Y, Z), the coordinates are (2, 3, 2) (FIG. 11). FIG. 12 shows a schematic flowchart of this processing, and FIG. 13 shows an example of the first processing for obtaining the X coordinate as a flowchart of each block search.

As described above, according to this embodiment, by the two cameras having the ND filters having different transmittances,
Good detection accuracy can be obtained even in a relatively low fault current region. In addition, by performing the 4-division processing three times in the place retrieval processing, 64-division retrieval can be performed as a result, and the processing time can be shortened.

Example 2. In the above embodiment, the retrieved block is 1 block out of 64 blocks, but as an actual phenomenon, the accident occurrence location may occur at the boundary of each block, and it is not possible to display only one block. However, it may be insufficient to determine the location of the accident. Therefore, in addition to the brightest block (hereinafter abbreviated as the confirmed block), an example in which candidate blocks are searched and displayed,
A second embodiment will be described with reference to FIGS. In the first embodiment, since the fixed block can be searched, the next brightest block should be searched for as a candidate block. In this example,
Next, set the parameter γ to the threshold 1 for searching the brightest block, and set the brightness L _{(X, Y, Z) of the} fixed block.
The threshold value 1 is obtained by comparing the threshold value with the brightness of the blocks above and below and to the left and right of the fixed block.
The brighter block is set as a candidate block and displayed as shown in FIG. FIG. 15 shows an example in which the upper / right blocks of the confirmed block are candidate blocks.
In this example, the upper right block of the fixed block sandwiched between the candidate blocks is displayed as a reference block, and further, it is handled so as not to be leaked. FIG. 14 shows a flowchart of the above processing.

As described above, according to this embodiment, the candidate block is searched and displayed in addition to the brightest block, so that the location of the accident can be accurately determined.

Example 3. In the above-described embodiment, since the number of cameras is increasing, there are side effects such as a long processing time of the system and a large image memory for recording image data.
Hereinafter, examples of a method for reducing the processing time and a method for reducing the memory will be described. A method for reducing the memory will be described with reference to FIGS. Generally, a television screen is called an interlace system, and by combining two field screens having a horizontal resolution of 512 pixels and a vertical resolution of 240 pixels, the horizontal resolution is 512 pixels and the vertical resolution is One frame screen image of 480 pixels is obtained.
As shown in FIG. 16, one frame is configured by combining an odd field and an even field. In the above-described embodiment, the location search is divided into eight parts in the vertical direction, and 60 scanning line segments constitute one block. Here, even if the vertical resolution is halved, one block
The number of scanning lines is 0, and the fact that the resolution is completely unaffected in noise removal processing is used in terms of detection accuracy. Here, as a method of recording in the image memory, the video data from one camera is assigned to the first to 240th scanning lines on the frame memory, and the video data from the other camera is assigned to the 241st to 480th lines. Allocate and record the same timing frame on the same frame memory (FIG. 17). As a result, since the time axes of the two cameras are the same in the frame search processing, only one of the cameras needs to be performed, and the processing time can be shortened by reducing the memory and the frame search processing time can be further reduced. Can be halved, and the processing time can be shortened to 1/4.

Example 4. Next, as an example of a method for reducing the processing time, Example 4 will be described with reference to FIG.
For shortened accidents or ground faults, take advantage of the fact that they always occur before the power failure signal that occurred as a result. That is, it can be said that the frame search process is performed without any waste from the frame image data at the time when the power failure signal is generated. FIG.
First, the sum of the brightness of the video data of the 11th frame is calculated, and the average value of the brightness is calculated (step 1
00-102). Next, the average value of the brightness of the 10th frame is similarly obtained, and the average value of the brightness of the 11th frame is calculated.
The sizes are compared (step 103). At this time, in order to remove an error due to a slight change in brightness, the determination is made based on whether or not there is a difference over a certain change width (denoted by α in this embodiment) (whether or not there is a decrease). . The same process is performed until the first frame until a change occurs. If no change is obtained even after processing up to the first frame, it is determined that no accident has occurred,
The result is displayed on the monitor screen, and the process ends. When a change occurs in the Nth frame, it is determined that the previous frame, that is, the N + 1th frame is the maximum brightness frame, and the result is stored as the T frame,
Move to the next location search process.

As described above, according to this embodiment, from the frame image data at the time when the power failure signal is generated,
The processing time can be shortened by performing the frame search processing backward.

Example 5. As a method of shortening the processing time, the fifth embodiment will be described with reference to FIG. Since the frame search processing is performed by utilizing the change in the average value of the brightness of the entire screen, it is possible to reduce the processing time by thinning the scanning lines as shown in FIG. That is, if the number of scanning lines to be processed is halved by thinning, the processing time is halved, and if the number of scanning lines is ¼, the processing time is ¼.

Example 6. In the fifth embodiment, an example of thinning the scanning lines has been described, but FIG. 20 shows a sixth embodiment in which the processing time is shortened by thinning out pixels in the horizontal direction.

Example 7. Further, in a system in which the accuracy of the generation time can be ignored, FIG. 21 shows an example as the eighth embodiment in which the processing time is shortened by thinning out the frames.

Example 8. In the above embodiments, the number of cameras used has been assumed to be two, but next, as an eighth embodiment, a method of shortening the processing time when the number of cameras exceeds two, for example, four, is described. An example will be described with reference to FIG. In this embodiment, in order to reduce the memory, as shown in FIG. 22, it is assumed that the frame memory for one screen is normally divided into four and allocated and used for each. Here, it is assumed that the transmittances of the ND filters attached to the four cameras have a power ratio. That is, the filter having the highest transmittance has a transmittance of 1/2, the filter having the second highest transmittance has a transmittance of 1/4, and the filters having the second highest transmittance have 1/8 and 1/16. With the above configuration, when starting the frame search processing, first, the frame data immediately after the power failure signal is generated is used, and the brightness of the area to which the camera with the transparency of 1/8 is allocated is first used. Calculate the average value of. If the result is compared with the preset average brightness of the optimum image, and if they are almost equal, then the temporal average brightness of the corresponding region of the frame one frame before is determined. Repeat to identify the accident occurrence frame. In reality, it is rare that the brightness is close to the average value of the optimum video, and the brightness is different in most cases. Therefore, if the average value of the brightness of the area to which the camera having the transparency of 1/8 is assigned is twice or more brighter than the average brightness of the optimum image, the transparency is 1/16. The process shifts to the process of obtaining the average value of the brightness of the regions to which the cameras are assigned, and conversely, if it is 1/2 or less, the average value of the brightness of the regions to which the camera of which the transparency is 1/4 is assigned is determined. When the processing shifts to 1/4 or less, the processing shifts to the processing of obtaining the average value of the brightness of the area to which the camera having the transparency of 1/2 is assigned.
The processing time can be shortened as compared with the processing for all the cameras.

[0060]

According to the first aspect of the present invention, there is provided an image processing apparatus having a plurality of ND filters having different transmissivities, for monitoring a short circuit / ground fault accident of electric power equipment, and an image storage means for recording image signals of the cameras. And the image analysis processing means for analyzing the image signal stored in the image storage means, the sensitivity range of the entire system can be expanded.

According to a second aspect of the present invention, in the image processing apparatus according to the first aspect, the first camera having a low transparency and the second camera having a high transparency are provided. The sensitivity range can be expanded.

According to a third aspect of the present invention, in the image processing apparatus according to the first aspect, the image storage means records the image signals before and after the power failure signal is generated, so that the processing time can be shortened.

According to a fourth aspect of the present invention, in the image processing apparatus according to the first aspect, the image analysis processing means specifies a frame in which an image in which a short circuit / ground fault has occurred is recorded, and which part of the specified frame is specified. Since it is searched whether is the brightest, image processing can be performed easily.

The image processing apparatus according to a fifth aspect of the present invention is the apparatus according to the first aspect, which is provided with a monitor device for displaying the result retrieved by the image analysis means, so that the image processing can be efficiently performed.

According to the image processing system of the sixth aspect, the number of frames set after the power failure signal is generated, and N which is different in transparency are set.
Image recording of image signals of a plurality of cameras having a D filter is continued, a frame in which a video in which an accident has occurred is recorded is specified from the captured image recording, and which part of the specified frame is brightest is searched. Since the retrieved result is displayed on the monitor screen, the sensitivity range of the entire system can be expanded.

The image processing system according to claim 7 is the image processing system according to claim 6.
In the above description, the area on the monitor screen is divided and the processing for searching for the brightest block is repeated to perform the location search processing, so that the processing time can be shortened.

The image processing system according to claim 8 repeats the process of dividing the 64 blocks of the area on the monitor screen into four and searching for the brightest block three times, so that the processing time can be shortened.

According to the image processing system of the ninth aspect, in addition to the brightest block, the candidate blocks in the vicinity are searched and displayed. Therefore, it is possible to search the location of the accident at the boundary.

According to the image processing system of the tenth aspect, a threshold value 1 for searching the brightest block next to the brightest fixed block is obtained by setting a parameter and multiplying by the brightness of the fixed block. The value is compared with the brightness of the upper, lower, left, and right blocks of the fixed block, and the block brighter than the threshold value 1 is set as the candidate block.

According to the image processing system of the eleventh aspect, the video data from the first camera is scanned by the scanning line 1 on the frame memory.
Since the first to predetermined lines are allocated, the video data from the second camera is allocated to the remaining scanning lines, and the same timing frame is recorded on the same frame memory, the memory amount can be reduced.

According to the twelfth aspect of the image processing system, the frame search processing is performed by tracing back the time from the frame image data at the time when the power failure signal is generated, so that the search processing time can be shortened.

In the image processing system according to the thirteenth aspect, since the frame search processing is performed by thinning out the data on the frame memory in the horizontal scanning line direction, the processing time can be shortened.

In the image processing system according to the fourteenth aspect, since the frame search processing is performed by thinning out the data on the frame memory in the pixel direction, the processing time can be shortened.

In the image processing system according to the fifteenth aspect, the frame search processing is performed by thinning out the data on the frame memory, so that the processing time can be shortened.

An image processing system according to a sixteenth aspect of the present invention has a plurality of cameras having an ND filter in which the transmittance is a power ratio, and a preset average value of the brightness at the time of the optimum image and among the cameras are set. The brightness of the areas to which one camera is assigned is compared, and the camera to be compared next is determined according to the result, so that the processing time can be shortened.

[Brief description of drawings]

FIG. 1 is a block diagram of an image processing system according to a first embodiment of the present invention.

FIG. 2 is an overall flowchart diagram (A) of the first embodiment of the present invention.

FIG. 3 is a flow chart (A) of image capturing according to the first embodiment of the present invention.

FIG. 4 is a frame search flowchart (A) according to the first embodiment of the present invention.

FIG. 5 is a location search flowchart (A) in the first embodiment of the present invention.

FIG. 6 is a diagram showing an example of a captured image according to the first embodiment of the present invention.

FIG. 7 is a relationship diagram of the brightness vs. sensor output in the first embodiment of the present invention.

FIG. 8 is a diagram showing an example of a captured image according to the first embodiment of the present invention.

FIG. 9 is a diagram showing an example of a captured image according to the first embodiment of the present invention.

FIG. 10 is a screen block division diagram according to the first embodiment of the present invention.

FIG. 11 is a location search screen diagram according to the first embodiment of the present invention.

FIG. 12 is a place search flowchart (B) in the first embodiment of the present invention.

FIG. 13 is a block diagram of a block search process according to the first embodiment of the present invention.

FIG. 14 is a flowchart of candidate block search processing according to the second embodiment of the present invention.

FIG. 15 is a diagram showing an example of a detection result display screen according to the second embodiment of the present invention.

FIG. 16 is a frame synthesis diagram according to the third embodiment of the present invention.

FIG. 17 is a memory reduction diagram according to the third embodiment of the present invention.

FIG. 18 is a frame search flowchart (B) according to the fourth embodiment of the present invention.

FIG. 19 is a diagram illustrating a scanning line thinning operation according to the fifth embodiment of the present invention.

FIG. 20 is a diagram illustrating a pixel thinning operation according to the sixth embodiment of the present invention.

FIG. 21 is a diagram illustrating a frame thinning operation according to the seventh embodiment of the present invention.

FIG. 22 is a diagram showing an example of screen composition in Example 8 of the present invention.

FIG. 23 is an overall flowchart diagram (C) of Embodiment 8 of the present invention.

FIG. 24 is a block diagram of an image processing system in Conventional Example 1.

FIG. 25 is a relationship diagram of brightness vs. sensor output in Conventional Example 1.

FIG. 26 is a diagram showing an example of image data in Conventional Example 1.

[Explanation of symbols]

1 Accident occurrence part, 2 Camera part, 2a 1st camera part, 2b 2nd camera part, 3 ND filter, 3a
1st ND filter, 3b 2nd ND filter, 4
Lens, 4a first lens, 4b second lens, 5
Microphone, 5a first microphone, 5b second microphone, 6
Camera body, 6a first camera body, 6b second camera body, 7 camera turntable, 7a first camera turntable, 7b second camera turntable, 8 monitor device, 9
Camera control signal line, 9a first camera control signal line,
9b second camera control signal line, 10 video signal transmission line, 10a first video signal transmission line, 10b second video signal transmission line, 11 video signal line, 12 image processing unit,
13 image processing unit built-in image processing apparatus, 14 image memory built-in image processing apparatus, 15 input operation unit, 16 input operation control signal line, sensitivity curve when 31 ND filter is not used, sensitivity when 32 ND filter is used Curve, 42 sensitivity curve when the first ND filter is used, 43 sensitivity curve when the second ND filter is used.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI Technical display location H02H 5/04 H H04N 7/18 D

Claims (16)

[Claims] 1. A plurality of cameras having ND filters having different transmittances for monitoring a short-circuit / ground fault accident of power equipment, an image storage means for recording image signals of the cameras, and the image storage means. And an image analysis processing unit that analyzes and processes the image signal. 2. The image processing apparatus according to claim 1, further comprising a first camera having low transparency and a second camera having high transparency. 3. The image processing apparatus according to claim 1, wherein the image storage means records the image signals before and after the power failure signal is generated. 4. The image analysis processing means specifies a frame in which an image in which a short circuit / ground fault has occurred is recorded, and searches which part of the specified frame is the brightest. The image processing device described. 5. The image processing apparatus according to claim 1, further comprising a monitor device for displaying a result retrieved by the image analysis means. 6. The image recording of the image signals of a plurality of cameras having ND filters having different set numbers of frames and transparency after the power failure signal is generated is continued, and an accident occurs from the captured image recording. An image processing system characterized in that a frame in which an image is recorded is specified, which part of the specified frame is brightest, and the search result is displayed on a monitor screen. 7. The image processing system according to claim 6, wherein an area on the monitor screen is divided and the processing for searching the brightest block is repeated to perform the location search processing. 8. A block on an area on the monitor screen is divided into 4 blocks.
The image processing system according to claim 7, wherein the process of dividing and searching for the brightest block is repeated three times. 9. The image processing system according to claim 7, wherein a candidate block in the vicinity of the brightest block is searched and displayed. 10. A threshold value 1 for searching a brightest block next to the brightest fixed block is obtained by setting a parameter and multiplying the brightness of the fixed block by the threshold value and the fixed block. 10. The image processing system according to claim 9, wherein a block that is brighter than a threshold value 1 is selected as a candidate block as compared with the brightness of the upper, lower, left, and right blocks. 11. The video data from the first camera is assigned to the first to predetermined scan lines on the frame memory, the video data from the second camera is assigned to the remaining scan lines, and the same timing frame is set. The image processing system according to claim 6, wherein the image is recorded on the same frame memory. 12. The image processing system according to claim 6, wherein the frame search processing is performed so as to go back in time from the frame image data at the time when the power failure signal is generated. 13. The image processing system according to claim 6, wherein the frame search process is performed by thinning out the data on the frame memory in the horizontal scanning line direction. 14. The image processing system according to claim 6, wherein the frame search process is performed by thinning out the data in the frame memory in the pixel direction. 15. The image processing system according to claim 6, wherein the frame search processing is performed by thinning out the data on the frame memory. 16. A plurality of cameras having an ND filter whose transmittance is a power ratio, wherein one of the cameras is assigned a preset average value of brightness during optimum video. 7. The image processing system according to claim 6, wherein the brightness of the area is compared, and the camera to be compared next is determined according to the result.