JP7187394B2 - Indicated value reading system, method and program - Google Patents

Description

The present invention relates to an indicated value reading system, method, and program for reading indicated values of various measuring instruments.

For example, in various plants, the operating states of various devices are measured by measuring instruments, and the operating states of the various devices are monitored based on the measured indicated values. In this case, the worker goes to the place where the equipment is installed and reads the indicated value of the measuring device provided for the equipment installed at the installation place. Then, the operator monitors and controls the operation of the equipment based on the indicated value read from the measuring instrument. Considering workability, it is considered to remotely read the indicated value of the measuring instrument. As such an indicated value reading system, for example, there is a system described in Patent Document 1 below. In the device status monitoring system described in Patent Document 1, a camera captures an image of a controlled device, a sensor collects information about the controlled device, transmits the captured image and the collected information to a remote monitoring control device, The monitoring and control device judges the state of the controlled device and displays a warning.

JP-A-2006-134063

In the device status monitoring system of Patent Document 1, an image of a controlled device captured by a camera and information about the controlled device collected by a sensor are sent to a remote monitoring control device, and the remote monitoring control device determines the status of the controlled device. do. In this case, the image and information of the controlled device are large-capacity data, and the image and information of the controlled device, which are large-capacity data, are transmitted to the remote monitoring control device. equipment is required. As a result, the remote monitoring system becomes complicated and large, and many devices and wirings need to be added to introduce the system, resulting in an increase in introduction cost.

An object of the present invention is to solve the above-described problems, and to provide an indicated value reading system, method, and program capable of suppressing an increase in introduction cost and easily reading the indicated value of a measuring instrument. .

An indicated value reading system of the present invention for achieving the above objects comprises a camera capable of photographing the pointer of a measuring instrument, and connected to the camera to process image data of the pointer photographed by the camera. The image processing device is characterized by comprising an image processing device that obtains an indicated value of the pointer, and a transmission device that transmits data of the indicated value of the pointer obtained by the image processing device.

Therefore, the camera captures the pointer of the measuring instrument, the image processing device processes the image data of the pointer to obtain the indicated value of the pointer, and the transmission device transmits the data of the indicated value of the pointer. Here, since the data of the indicated value of the indicator transmitted by the transmitting device is a numerical value as digital data, it is a low-capacity data, and a transmitting device for high-speed transmission of large-capacity data is not required. As a result, it is possible to suppress an increase in introduction cost and to easily read the indicated value of the measuring instrument.

The indicated value reading system of the present invention is characterized in that the camera and the image processing device are connected by wire.

Therefore, since the camera and the image processing device are connected by wire, the device can be simplified by eliminating the need for a wireless device.

In the indicated value reading system of the present invention, the image processing device processes the image data of the pointer to obtain the indicated value of the pointer, and then deletes the image data of the pointer.

Therefore, after the image processing device processes the image data of the pointer and obtains the indication value, the image data of the pointer is deleted. The device can be simplified by eliminating the need for a large storage device.

In the indicated value reading system of the present invention, the operating range and unit of the indicating needle are assigned to the measuring instrument, and the image processing device processes the image data of the indicating needle to obtain the indicated value of the indicating needle. It is characterized by obtaining a numerical value as

Therefore, since the operation range and unit of the pointer are assigned, the image processing device can easily obtain the numerical value as the indicated value of the pointer by processing the image data of the pointer.

The indicated value reading system according to the present invention is characterized in that a storage unit is provided for storing the indicated value of the pointer obtained by the image processing device.

Therefore, since the indicated value of the indicator obtained by the image processing device is stored in the storage unit, even if there is a failure in data transmission by the transmitting device, the data of the indicated value of the indicator can be continuously obtained. .

In the indicated value reading system of the present invention, the image processing device obtains the likelihood of the indicated value of the indicated needle obtained by processing the image data of the indicated needle, and the likelihood is lower than a preset threshold value. In some cases, the image data of the pointer is stored in the storage unit.

Therefore, when the likelihood of the indication value of the pointer is low, the image data of the pointer obtained by obtaining the indication value with the low likelihood is stored in the storage unit. It is possible to identify the cause of the lower likelihood of the indicated value.

In the indicated value reading system of the present invention, the image processing device obtains the indicated value of the pointer at each predetermined processing cycle, and the transmitting device obtains data of the indicated value of the pointer at each predetermined transmission cycle. When the likelihood is lower than the threshold value, the transmission device transmits the data of the indication value of the pointer obtained last time by the image processing device.

Therefore, when the likelihood of the indication value of the pointer is low, the data of the indication value of the pointer can be continuously obtained because the transmission device transmits the data of the indication value of the pointer obtained last time.

In the indicated value reading system of the present invention, the image processing device obtains the likelihood of the indicated value of the indicated needle obtained by processing the image data of the indicated needle, and the likelihood is a preset first threshold value. If it is equal to or greater than the above, while transmitting the data of the indication value of the indicator by the transmitting device, if the likelihood is equal to or greater than a second threshold lower than the first threshold and smaller than the first threshold, the indicator is corrected by the similarity value or the previous value of the image collation and transmitted by the transmission device.

Therefore, when the likelihood is lower than the first threshold and equal to or higher than the second threshold, the indication value of the pointer is corrected by the similar value or the previous value and transmitted, so the accuracy of the indication value of the pointer to be transmitted can be improved.

In the indicated value reading system of the present invention, the likelihood is lower than the first threshold value and equal to or higher than the second threshold value, and the absolute value of the indicated value of the pointer and the similarity value is determined in advance. If it is equal to or less than a set deviation threshold value, the indicated value of the pointer is corrected by the similarity value and transmitted by the transmitting device.

Therefore, if the absolute value between the indicated value of the indicator and the similarity value is less than the deviation threshold, the indicated value of the indicator is corrected by the similarity value and transmitted, improving the accuracy of the indicated value of the indicator to be transmitted. can do.

In the indicated value reading system of the present invention, the image processing device has an absolute value between the indicated value of the pointer and the similarity value higher than the deviation threshold, the similarity of the similarity value is equal to or higher than a third threshold, When the absolute value of the indicated value of the pointer and the similarity value is equal to or less than the absolute value of the previous value and the similarity value, the indicated value of the pointer is corrected by the similarity value and transmitted by the transmitting device. characterized by

Therefore, when the degree of similarity is equal to or higher than the third threshold and the absolute value between the indicator value and the similarity value is equal to or smaller than the absolute value between the previous value and the similarity value, the indicator value is corrected by the similarity value. Therefore, the accuracy of the transmitted indicator value can be improved.

In the indicated value reading system of the present invention, the image processing device has an absolute value between the indicated value of the pointer and the similarity value higher than the deviation threshold, the similarity of the similarity value is equal to or higher than a third threshold, When the absolute value of the indicated value of the pointer and the similarity value is higher than the absolute value of the previous value and the similarity value, the indicated value of the pointer is corrected by the previous value and transmitted by the transmitting device. It is characterized by

Therefore, when the degree of similarity is equal to or higher than the third threshold and the absolute value between the indicator value and the similarity value is higher than the absolute value between the previous value and the similarity value, the indicator value is corrected by the previous value. Since the information is transmitted, the accuracy of the indication value of the indicator needle to be transmitted can be improved.

In the indicated value reading system of the present invention, the camera can photograph a plurality of pointers of the measuring instrument, and the image processing device sequentially processes the image data of the plurality of pointers photographed by the camera. to obtain the indicated value of the pointer.

Therefore, since the image processing device sequentially processes the image data of a plurality of pointers to obtain the indicated values of the pointers, it is possible to simplify the device by eliminating the need for a plurality of cameras.

In the indicated value reading system of the present invention, the image processing device processes image data captured by the camera to create image data of the measuring instrument, and processes the image data of the measuring instrument using deep learning. is characterized in that the indicated value of the pointer in is obtained.

Therefore, since the indicated value of the pointer is obtained by deep-learning the image data of the pointer, the indicated value of the pointer can be quickly obtained with high accuracy.

In the indicated value reading system of the present invention, a monitor can be connected to the image processing device, and the image data of the camera and the image data of the measuring instrument processed by the image processing device can be displayed on the monitor. characterized by being

Therefore, since the image data of the camera and the image data of the measuring device processed by the image processing device are displayed on the monitor, the operator can confirm the visual angle of the camera and the specified measuring device.

In the indicated value reading system of the present invention, the image processing device is capable of displaying setting data necessary for processing and indicated value data of the pointer obtained by the image processing device on the monitor. Characterized by

Therefore, the setting data necessary for processing and the indicated value data of the indicator obtained by the image processing device are displayed on the monitor. can be confirmed.

The indicated value reading method of the present invention includes the steps of photographing a pointer of a measuring instrument, processing the photographed image data of the pointer to obtain the indicated value of the pointer, and reading the image data of the pointer. a step of deleting and transmitting the obtained data of the indication value of the pointer.

Therefore, since the data of the indicated value of the indicator transmitted by the transmitting device is a numerical value as digital data, it becomes low-capacity data, and a transmitting device for high-speed transmission of large-capacity data is not required. As a result, it is possible to suppress an increase in introduction cost and to easily read the indicated value of the measuring instrument.

The indicated value reading program of the present invention includes a process of photographing the pointer of the measuring instrument, a process of processing the photographed image data of the pointer to obtain the indicated value of the pointer, and a process of obtaining the indicated value of the pointer. and a process of deleting and transmitting the obtained data of the indication value of the pointer.

Therefore, since the data of the indicated value of the indicator transmitted by the transmitting device is numerical value as digital data, it becomes low-capacity data, and a large-sized transmitting device for transmitting large-capacity data is not required. As a result, it is possible to suppress an increase in introduction cost and to easily read the indicated value of the measuring instrument.

ADVANTAGE OF THE INVENTION According to the indicated value reading system, method, and program of the present invention, it is possible to suppress an increase in introduction cost and easily read indicated values of measuring instruments.

FIG. 1 is a schematic configuration diagram showing an indicated value reading system of this embodiment. FIG. 2 is a schematic diagram showing a monitor screen in the indicated value reading system. FIG. 3 is a flow chart showing a method of reading indicated values. FIG. 4 is a flow chart showing the instruction value estimation method. FIG. 5 is a graph for explaining the degree of similarity. FIG. 6 is a schematic diagram for explaining a method of correcting indicated values.

Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the present invention is not limited by this embodiment, and when there are a plurality of embodiments, the present invention includes a combination of each embodiment.

FIG. 1 is a schematic configuration diagram showing an indicated value reading system of the present embodiment, and FIG. 2 is a schematic perspective view showing a monitor image in the indicated value reading system.

In this embodiment, as shown in FIG. 1, an indicated value reading system 10 is arranged at a site 100 where various devices (not shown) are installed. The indicated value reading system 10 includes at least a camera 11 , an image processing device 12 and a transmission device 13 . The indicated value reading system 10 includes a plurality of (six in this embodiment) measuring instruments 51, 52, 53, 54, 55, and 56 of various devices arranged at the site 100. Indicator needles 61, 62, 63, It reads the indicated values of 64, 65 and 66 and transmits them to the monitoring system 20 in the management center 200. FIG. The management center 200 is separated from the site 100 and monitors various devices arranged at the site 100 by the monitoring system 20 . Note that the number of measuring instruments is not limited to six.

Here, the measuring instruments 51, 52, 53, 54, 55, and 56 are thermometers, pressure gauges, flowmeters, and the like. Measuring instruments 51, 52, 53, 54, 55, and 56 have a circular instrument panel, pointers 61, 62, 63, 64, 65, and 66 are rotatably supported at the center, and a scale 71 is provided around the periphery. , 72, 73, 74, 75 and 76 are attached. The pointers 61, 62, 63, 64, 65, and 66 rotate according to the measured values, and point the tips to the scales 71, 72, 73, 74, 75, and 76 to display the indicated values.

The indicated value reading system 10 is arranged at a site 100, and specifically includes a camera 11, an image processing device 12, a transmission device 13, a storage section 14, an operation section 15, a monitor 16, and a power supply section 17. and The monitoring system 20 is located at the management center 200 and comprises a control device 21 , a receiving device 22 and a monitor 23 .

The camera 11 can photograph pointers 61 , 62 , 63 , 64 , 65 , 66 of a plurality of measuring instruments 51 , 52 , 53 , 54 , 55 , 56 . The camera 11 can photograph a predetermined area including the measuring instruments 51 , 52 , 53 , 54 , 55 and 56 . Therefore, by specifying the area in which each of the measuring instruments 51, 52, 53, 54, 55, and 56 is arranged in the shooting area of the camera 11, the measuring instruments 51, 52, 53, 54, 55, and 56 are specified. do. In this case, the operator can designate only a specific measuring instrument among the plurality of measuring instruments 51, 52, 53, 54, 55, and 56 by using the program stored in the operating section 15 or the storage section 14. FIG.

The image processing device 12 is wired to the camera 11, but the image processing device 12 and the camera 11 may be wirelessly connected. The image processing device 12 has an image processing section 12a, an image recognition section 12b, and a display creation section 12c. Here, the image processing device 12 is, for example, a CPU (Central Processing Unit) and a memory. The image processing unit 12 a preprocesses image data of the pointers 61 , 62 , 63 , 64 , 65 , and 66 captured by the camera 11 . Specifically, the image processing unit 12a performs teacher data creation processing and learning processing, filtering processing of image data captured by the camera 11, and binarization processing of image data captured by the camera 11. FIG.

The image processing unit 12a obtains indicated values of the pointers 61, 62, 63, 64, 65, 66 by performing image recognition by deep learning as described later. For example, a large number of sample data of the pointers 61, 62, 63, 64, 65, and 66 to be processed are acquired in advance, and data for teaching is created by learning the sample data. The image recognition unit 12b compares the image data of the pointers 61, 62, 63, 64, 65, 66 photographed and image-processed by the camera 11 with teacher data.

Further, the image data captured by the camera 11 is image data of a predetermined area including the measuring instruments 51, 52, 53, 54, 55, and 56. FIG. The image processing unit 12a extracts only the image data of the measuring instruments 51, 52, 53, 54, 55, and 56 from the image data captured by the camera 11. FIG. Furthermore, the image processing unit 12a binarizes the image data of the pointers 61, 62, 63, 64, 65, and 66 captured by the camera 11, are black, and the instrument panels of the measuring instruments 51, 52, 53, 54, 55, 56 are white. Note that the image processing of each pointer 61, 62, 63, 64, 65, 66 is not limited to the binarization process, and the binarization process may be omitted even if it is another process.

As described above, the image recognition unit 12b compares the image data of the pointers 61, 62, 63, 64, 65, 66 photographed by the camera 11 and image-processed by the image processing unit 12a with data for teachers. do. By collating this image data, the shapes of the pointers 61, 62, 63, 64, 65, and 66 are specified, and the position pointed by the tip is specified to obtain the indicated value. In this embodiment, the indicated values of pointers 61, 62, 63, 64, 65, and 66 in measuring instruments 51, 52, 53, 54, 55, and 56 are obtained by deep learning.

A kernel (filter) B, which is grid-like numerical data, is prepared in advance, and the image data A of each of the measuring instruments 51, 52, 53, 54, 55, and 56 processed and binarized by the image processing unit 12a and this kernel B is used for image processing. First, the numerical data of the partial image (window) in the kernel B and the image data A of the same size as the kernel B are subjected to a product-sum operation in which the elements are multiplied by each pixel and then added to sum. Calculate the numerical value of Next, the range of the partial image in the image data A is shifted by several pixels, and the sum-of-products operation is performed in the same manner as described above to calculate the second numerical value. Subsequently, this process is repeated until the image data A is covered, and folding is performed to obtain reduced lattice data smaller than the image data A, that is, the feature map C. FIG. Then, with the feature map C as an input, deep learning (learning by a neural network) is performed to estimate the indicated values of the pointers 61, 62, 63, 64, 65, and 66, and to calculate the likelihood. Here, deep learning in which a convolution operation is introduced into a neural network is used, but deep learning in which a pooling operation is introduced into a neural network, for example, may be used. Further, the image processing by the image processing device 12 is not limited to deep learning, and other machine learning or other analysis methods may be used.

Note that the camera 11 acquires a plurality of measuring instruments 51, 52, 53, 54, 55, and 56 as moving images, and the image processing device 12 acquires still images from the moving images acquired by the camera 11 at predetermined timings, and Still images of the pointers 61, 62, 63, 64, 65, 66 are processed to obtain the indicated values of the respective pointers 61, 62, 63, 64, 65, 66. In this case, the image processing device 12 sets the operation range (scale 71, 72, 73, 74, 75, 76), numerical values of scales 71, 72, 73, 74, 75, 76, and units of measured values. Note that the timing at which the image processing device 12 captures a still image depends on the processing capability of the image processing device 12 .

Therefore, the image processing device 12 processes the image data of the pointers 61, 62, 63, 64, 65, and 66 captured by the camera 11 so that the pointers 61, 62, 63, 64, 65, and 66 Numerical values as indicated values for the pointers 61, 62, 63, 64, 65 and 66 are obtained based on the positions of and the initial set values.

The image recognition unit 12b estimates the indication values of the pointers 61, 62, 63, 64, 65, and 66 by deep learning and calculates the likelihood. is obtained, and the indication value with the highest likelihood is adopted. Then, the image recognition unit 12b processes the image data of the pointers 61, 62, 63, 64, 65, and 66 to obtain the indicated values adopted by the pointers 61, 62, 63, 64, 65, and 66. is compared with a preset threshold. Here, when the likelihood of the indication value of each pointer 61, 62, 63, 64, 65, 66 is lower than a preset threshold value, the image recognition unit 12b The image data for which the indicated values of 63, 64, 65, and 66 are obtained are stored in the storage unit 14 . Here, the likelihood of the indication value is the image matching rate by deep learning, and the threshold is, for example, 90%, but the numerical value of the threshold may be set appropriately.

The image processing apparatus 12 also has a storage unit (internal storage) (not shown), which is, for example, a RAM (Random Access Memory). The storage unit of the image processing device 12 stores an indicated value reading program for the image processing device 12 to process the image data of the pointers 61, 62, 63, 64, 65, and 66 to obtain the indicated values. . That is, data for the image processing device 12 to perform deep learning of image data is stored. The indicated value reading program is a process of photographing the pointers 61, 62, 63, 64, 65, and 66 of the measuring instruments 51, 52, 53, 54, 55, and 56; , 65 and 66 are processed to obtain indicated values of the pointers, the image data of the pointers 61, 62, 63, 64, 65 and 66 are deleted, It causes the image processing apparatus (computer) 12 to execute a process of transmitting the data of the indicated values 63, 64, 65, and 66. FIG.

The image processing device 12 obtains the indicated values of the pointers 61, 62, 63, 64, 65 and 66 at predetermined intervals. Data of indicated values of 65 and 66 are transmitted. The predetermined period of transmission by the transmitter 13 depends on the transmission capability. Here, it is not necessary to match the predetermined cycle for image processing by the image processing device 12 and the predetermined cycle for transmitting the data of the instruction value by the transmission device 13 . Then, when the image recognition unit 12b determines that the likelihood of the image data of each pointer 61, 62, 63, 64, 65, 66 is low, the transmission device 13 detects the pointer 61 obtained by the image processing device 12 last time. , 62, 63, 64, 65, and 66 are transmitted again.

The display creation unit 12c creates an image to be displayed on the monitor 16. FIG. The image created by the display creating unit 12c and displayed on the monitor 16 will be described later.

Note that when the camera 11 captures the pointers 61 , 62 , 63 , 64 , 65 , 66 of the plurality of measuring instruments 51 , 52 , 53 , 54 , 55 , 56 , the image processing device 12 , 62, 63, 64, 65 and 66 are processed in a preset order to obtain the indicated values of the pointers 61, 62, 63, 64, 65 and 66 respectively.

The transmission device 13 transmits the data of the indication values of the pointers 61 , 62 , 63 , 64 , 65 and 66 obtained by the image processing device 12 . The data transmitted by the transmitting device 13 are the date and time, the indicated value (numerical value), and the likelihood. At this time, the image processing device 12 processes the image data of the pointers 61, 62, 63, 64, 65 and 66 to obtain the indicated values of the pointers 61, 62, 63, 64, 65 and 66. , and the image data of the pointers 61, 62, 63, 64, 65, and 66 that follow are fetched and updated. That is, the image processing device 12 does not store the image data of the pointers 61, 62, 63, 64, 65, 66, and the transmission device 13 stores the image data of the pointers 61, 62 processed by the image processing device 12. , 63, 64, 65, and 66 are not transmitted, but only the data of the obtained indication values of the pointers 61, 62, 63, 64, 65, and 66 are transmitted. Here, the data of the indicated values of the pointers 61, 62, 63, 64, 65 and 66 are the date and time of processing, the numerical values of the measuring instruments 51, 52, 53, 54, 55 and 56, the data of the pointers 61 , 62, 63, 64, 65, and 66.

The storage unit 14 is connected to the image processing device 12 . The storage unit 14 is, for example, a ROM (Read Only Memory). When the image processing device 12 obtains the indication values of the pointers 61, 62, 63, 64, 65, and 66, the storage unit 14 stores the indication value data transmitted by the transmission device 13. FIG. In addition, as described above, the storage unit 14 stores in the storage unit 14 the image data obtained by obtaining the indicated values of the pointers 61, 62, 63, 64, 65, and 66 with the low likelihood by the image recognition unit 12b. .

The operation unit 15 is used by an operator to input various setting data to the image processing apparatus 12 . Here, the various setting data include the date and time, the shooting area of the camera 11, the measuring instruments 51, 52, 53, 54, 55, and 56 for image processing, the data storage destination, and the like.

The monitor 16 is provided with a plurality of (four in this embodiment) display sections 31, 32, 33, and 34, as shown in FIG. The first display unit 31 displays all image data of the area captured by the camera 11 . The second display unit 32 displays the images of the measuring instruments 51, 53, 54 of the specific indicator needles 61, 63, 64 whose indicated values are to be read, among the plurality of measuring instruments 51, 52, 53, 54, 55, 56. indicate. The third display unit 33 displays setting data required for processing by the image processing device 12 . This setting data includes the above-mentioned date and time, the shooting area of the camera 11, the measuring instruments 51, 52, 53, 54, 55, and 56 for image processing, the data storage destination, and the processing state of the image processing device 12. indicate. The fourth display unit 34 displays the binarized image of the pointer 61 of the measuring device 51 obtained by the image processing device 12 and the indicated value (numerical value) of the pointer 61 .

The power supply unit 17 supplies power to the camera 11 , the transmission device 13 , the storage unit 14 , the operation unit 15 , the monitor 16 and the like via the image processing device 12 .

On the other hand, in the monitoring system 20, the receiving device 22 can be connected to the transmitting device 13 of the indicated value reading system 10 via a wireless network capable of transmitting and receiving low-capacity data. The receiving device 22 receives the indicated values of the pointers 61 , 62 , 63 , 64 , 65 , 66 transmitted by the transmitting device 13 of the indicated value reading system 10 . The control device 21 receives the indicated values of the pointers 61 , 62 , 63 , 64 , 65 and 66 received by the receiving device 22 and displays them on the monitor 23 . The control device 21 is connected to, for example, a monitoring device (not shown) and monitors various devices based on the indicated values of the indicators 61 , 62 , 63 , 64 , 65 and 66 .

Here, the instruction value reading method of the present embodiment will be specifically described. FIG. 3 is a flow chart showing a method of reading indicated values.

The indicated value reading method of the present embodiment includes a step of photographing the pointers 61, 62, 63, 64, 65, and 66 of the measuring instruments 51, 52, 53, 54, 55, and 56; , 63, 64, 65, 66 to obtain the indicated values of the pointers 61, 62, 63, 64, 65, 66; a step of deleting the image data and transmitting the obtained data of the indicated values of the pointers 61, 62, 63, 64, 65, 66;

Specifically, as shown in FIGS. 1 and 3, in step S11, the image processing device 12 receives the instructions of the plurality of measuring instruments 51, 52, 53, 54, 55, and 56 photographed by the camera 11. Image data of the needles 61, 62, 63, 64, 65, 66 are acquired. In this case, if only a specific measuring instrument among the plurality of measuring instruments 51, 52, 53, 54, 55, and 56 is specified, only the image data of the specified pointer is obtained. In step S12, the image processing device 12 image-processes the acquired image data of the pointers 61, 62, 63, 64, 65, and 66 of the plurality of measuring instruments 51, 52, 53, 54, 55, and 56. .

In step S13, the image processing device 12 performs deep learning on the image data of the pointers 61, 62, 63, 64, 65, and 66 of the measuring instruments 51, 52, 53, 54, 55, and 56 that have undergone image processing. do. In step S14, the indicated values and likelihoods of the pointers 61, 62, 63, 64, 65 and 66 are obtained. Then, in step S15, after performing deep learning, the image processing device 12 determines whether the likelihood of the obtained indication values of the pointers 61, 62, 63, 64, 65, and 66 is higher than the threshold. . Here, if the likelihood is higher than the threshold (Yes), the process proceeds to step S16.

Then, in step S16, data of indicated values (numerical values) of the pointers 61, 62, 63, 64, 65 and 66 are stored in the storage unit 14. FIG. In step S17, the image data of the pointers 61, 62, 63, 64, 65 and 66 of the measuring instruments 51, 52, 53, 54, 55 and 56 are updated and deleted. In step S<b>18 , the transmission device 13 transmits the data of the indicated values (numerical values) of the pointers 61 , 62 , 63 , 64 , 65 , 66 to the monitoring system 20 . On the other hand, if the likelihood of the indicated values of the pointers 61, 62, 63, 64, 65, and 66 obtained in step S15 is lower than the threshold (No), in step S19, each pointer 61 with low likelihood , 62, 63, 64, 65, and 66 are stored in the storage unit 14.

In the above process, when it is determined in step S15 that the likelihood is lower than the threshold value, the image data obtained by obtaining the indication values of the pointers 61, 62, 63, 64, 65, and 66 with low likelihood is stored in the storage unit 14, but the processing is not limited to this. FIG. 4 is a flow chart showing the indicator value estimation method, FIG. 5 is a graph for explaining the similarity, and FIG. 6 is a schematic diagram for explaining the indicator value correction method.

As shown in FIG. 4, in step S21, it is determined whether or not the likelihood s of the indicating needles 61, 62, 63, 64, 65, and 66 at the indicated value x is equal to or greater than a first threshold value α1 (for example, 0.8). do. Here, if the likelihood s is equal to or greater than the first threshold value α1 (Yes), the calculated indication value x of the pointers 61, 62, 63, 64, 65, and 66 is output in step S22. On the other hand, if the likelihood s is lower than the first threshold α1 (No), it is determined in step S23 whether the likelihood s is equal to or greater than a second threshold α2 (for example, 0.5). The second threshold value α2 is a value lower than the first threshold value α1.

Here, if the likelihood s is equal to or greater than the second threshold value α2 (Yes), in step S24, it is determined whether the absolute value of the deviation between the instruction value x and the similarity value z is equal to or less than a preset deviation threshold value β1. judge. Here, the similarity value z is the highest matching probability (similarity degree u). As shown in FIG. 5, for example, as a result of comparing the image data of pointers 61, 62, 63, 64, 65, and 66 with the teacher data, the probability (similarity Since u) is the highest, let it be the similarity value z.

If the absolute value of the deviation between the indicated value x and the similarity value z is less than the deviation threshold value β1 (Yes), then in step S25 the calculated indicated values of the pointers 61, 62, 63, 64, 65, and 66 A correction value F (x, z, s, u) obtained by correcting x is output. Here, a method for correcting the indicated value x will be described. As shown in FIG. 6, when two estimated values a and b (for example, indicated value x and similarity value z) are obtained, a correction value F(a,b,s,t) is calculated. Specifically, the indicated value F (a, b, c, d) corrected by the following formula is calculated.
F(a,b,c,d)=a−{(a−b)*d/(c+d)}
where a and b are estimated values and c and d are coefficients. That is, the correction value F (x, z, s, u) obtained in step S25 uses the indicated value x and similarity value z for a and b, and the likelihood s and similarity u for c and d. is used.

On the other hand, if the absolute value of the deviation between the indicated value x and the similarity value z is greater than the deviation threshold value β1 in step S25 (No), in step S26, the similarity u with respect to the similarity value z is set to the third threshold value α3 (for example, , 0.5). The third threshold value α3 is preferably the same value as the second threshold value α2, but may be a different value.

Here, if the degree of similarity u is equal to or greater than the third threshold value α3 (for example, 0.5) (Yes), then in step S27, the deviation between the indicated value x and the similarity value z is the previous indicated value (hereinafter referred to as the previous value) is less than or equal to the deviation between y and the similarity value z. Here, if the deviation between the indicated value x and the similarity value z is less than or equal to the deviation between the previous value y and the similarity value z (Yes), then in step S25, the calculated pointers 61, 62, 63, 64, and 65 , 66, and outputs an indicated value F(x, z, s, u) obtained by correcting the indicated value x. On the other hand, if the deviation between the indicated value x and the similarity value z is greater than the deviation between the previous value y and the similarity value z (No), then in step S28, the calculated pointers 61, 62, 63, 64, and 65 , 66, and outputs a correction value F(y, z, t, u) obtained by correcting the indicated value x. t is a constant (set value). The indicated value F (y, z, t, u) is calculated by the above-described formula, and the previous value y and the similarity value z are used for a and b, and the constant t and the similarity u are used for c and d. is used.

If the degree of similarity u is lower than the third threshold α3 (No) in step S26, the previous value y is output as an instruction value in step S30. Further, when the likelihood s is lower than the second threshold α2 (No) in step S23, it is determined in step S29 whether or not the degree of similarity u is equal to or greater than a fourth threshold α4 (for example, 0.5). The fourth threshold value α4 is preferably the same value as the second threshold value α2, but may be a different value. Here, if the degree of similarity u is equal to or greater than the fourth threshold value α4 (Yes), then in step S28, the correction value F ( , z, t, u). On the other hand, if the degree of similarity u is lower than the fourth threshold α4 (No), the previous value y is output as the instruction value in step S30.

As described above, the indicated value reading system of this embodiment includes the camera 11 capable of photographing the pointers 61, 62, 63, 64, 65, and 66 of the measuring instruments 51, 52, 53, 54, 55, and 56, and the camera 11 connected to the camera 11 to process the image data of the pointers 61, 62, 63, 64, 65, 66 and obtain the indicated values of the pointers 61, 62, 63, 64, 65, 66. 12 and a transmission device 13 for transmitting the data of the indication values of the pointers 61 , 62 , 63 , 64 , 65 , 66 obtained by the image processing device 12 .

Therefore, the camera 11 photographs the pointers 61, 62, 63, 64, 65, 66 of the measuring instruments 51, 52, 53, 54, 55, 56, and the image processing device 12 detects the pointers 61, 62, 63, 64. , 65 and 66 are processed to obtain the indicated values of the pointers 61, 62, 63, 64, 65 and 66. data. Here, since the data of the indicated values of the pointers 61, 62, 63, 64, 65, and 66 transmitted by the transmitting device 13 are numerical values as digital data, they are low-capacity data, and large-capacity data are transmitted at high speed. No transmitter is required. As a result, an increase in introduction cost can be suppressed, and the indicated value of the measuring instrument can be easily read.

In the indicated value reading system of the present embodiment, a camera 11 and an image processing device 12 are wired. Therefore, the device can be simplified by eliminating the need for a wireless device.

In the indicated value reading system of the present embodiment, the image processing device 12 processes the image data of the pointers 61, 62, 63, 64, 65 and 66 to After obtaining the indicated values, the image data of the pointers 61, 62, 63, 64, 65 and 66 are deleted. Therefore, there is no need to store the image data of the pointers 61, 62, 63, 64, 65, 66, which are large-capacity data, and the device can be simplified by eliminating the need for a large storage device.

In the indicated value reading system of the present embodiment, the operation ranges and units of the pointers 61, 62, 63, 64, 65, and 66 are assigned to the measuring instruments 51, 52, 53, 54, 55, and 56, and the image The processing device 12 processes the image data of the pointers 61, 62, 63, 64, 65 and 66 to obtain numerical values as the indicated values of the pointers 61, 62, 63, 64, 65 and 66. FIG. Therefore, the image processing device 12 processes the image data of the pointers 61, 62, 63, 64, 65 and 66 to easily obtain numerical values as the indicated values of the pointers 61, 62, 63, 64, 65 and 66. can ask.

In the indicated value reading system of this embodiment, a storage unit 14 is provided for storing the indicated values of the pointers 61, 62, 63, 64, 65, and 66 obtained by the image processing device 12. FIG. Therefore, since the indicated values of the pointers 61, 62, 63, 64, 65, and 66 obtained by the image processing device 12 are stored in the storage unit 14, even if there is a data transmission failure by the transmission device 13, the pointers The data of indicated values of 61, 62, 63, 64, 65 and 66 can be obtained continuously.

In the indicated value reading system of the present embodiment, the image processing device 12 processes the image data of the pointers 61, 62, 63, 64, 65, and 66 to obtain the pointers 61, 62, 63, 64, 65, The indicator 61, 62, 63, 64, 65, and 66, which have low likelihoods when the likelihood is lower than a preset threshold value, are obtained. The image data of 62, 63, 64, 65 and 66 are stored in the storage unit 14. FIG. Therefore, based on the image data of the pointers 61 , 62 , 63 , 64 , 65 , 66 stored in the storage unit 14 , it is possible to identify the cause of the low likelihood of the indicated value.

In the indicated value reading system of this embodiment, the image processing device 12 obtains the indicated values of the pointers 61, 62, 63, 64, 65, and 66 at predetermined intervals, and the transmitting device 13 reads the indicated values from the pointers at predetermined intervals. 61, 62, 63, 64, 65, and 66 are transmitted, and the image recognition unit 12b determines that the likelihood of the indication values of the pointers 61, 62, 63, 64, 65, and 66 is lower than the threshold. At this time, the transmission device 13 transmits the data of the indication values of the pointers 61, 62, 63, 64, 65, and 66 obtained by the image processing device 12 last time. Therefore, the data of the indicated values of the pointers 61, 62, 63, 64, 65, 66 can be continuously obtained.

In the indicated value reading system of the present embodiment, the image processing device 12 obtains the likelihood s of the indicating needles 61, 62, 63, 64, 65, and 66 at the indicated value x, and the first If it is equal to or greater than the threshold value α1, the data of the indicated value x of the pointers 61, 62, 63, 64, 65, and 66 is transmitted. , corrected by the similar value z or the previous value y, and transmitted. Therefore, it is possible to improve the accuracy of the indicated values of the pointers 61, 62, 63, 64, 65, 66 to be transmitted.

In the indicated value reading system according to the embodiment, the image processing device 12 determines that the likelihood s is lower than the first threshold value α1 and equal to or greater than the second threshold value α2, and the pointers 61, 62, 63, 64, 65, and 66 indicate When the absolute value of the value x and the similarity value z is equal to or less than a preset deviation threshold value β1, the indication value x of the pointers 61, 62, 63, 64, 65, and 66 is corrected by the similarity value z and transmitted. . Therefore, it is possible to improve the accuracy of the indicated value of the pointer to be transmitted.

In the indicated value reading system of the present embodiment, the image processing device 12 determines that the absolute values of the indicated value x and the similarity value z of the pointers 61, 62, 63, 64, 65, and 66 are higher than the deviation threshold β1, and the similarity u is greater than or equal to the third threshold value α3, and the absolute values of the indicated value x and the similarity value z of the pointers 61, 62, 63, 64, 65, and 66 are less than or equal to the absolute values of the previous value y and the similarity value z. , the indication value x of the pointers 61, 62, 63, 64, 65, 66 is corrected by the similarity value z and transmitted. Therefore, it is possible to improve the accuracy of the indicated value of the pointer to be transmitted.

In the indicated value reading system of the present embodiment, the image processing device 12 determines that the absolute values of the indicated value x and the similarity value z of the pointers 61, 62, 63, 64, 65, and 66 are higher than the deviation threshold β1, and the similarity u is greater than or equal to the third threshold value α3, and the absolute values of the indicated value x and the similarity value z of the pointers 61, 62, 63, 64, 65, and 66 are higher than the absolute values of the previous value y and the similarity value z. , the indicated value x of the pointers 61, 62, 63, 64, 65 and 66 is corrected by the previous value y and transmitted. Therefore, it is possible to improve the accuracy of the indicated value of the pointer to be transmitted.

In the indicated value reading system of this embodiment, the camera 11 can photograph pointers 61, 62, 63, 64, 65, and 66 of a plurality of measuring instruments 51, 52, 53, and 54, and the image processing device 12 , the image data of the pointers 61, 62, 63, 64, 65 and 66 photographed by the camera 11 are sequentially processed to obtain the indicated values of the pointers 61, 62, 63, 64, 65 and 66. Therefore, it is possible to simplify the device by eliminating the need for a plurality of cameras 11 .

In the indicated value reading system of the present embodiment, the image processing device 12 processes the image data captured by the camera 11 to create image data of the measuring instruments 51, 52, 53, 54, 55, and 56, and Obtain the indicated values of pointers 61, 62, 63, 64, 65, and 66 in measuring instruments 51, 52, 53, 54, 55, and 56 by deep learning of the image data of 51, 52, 53, 54, 55, and 56. . Therefore, highly accurate indicated values of the pointers 61, 62, 63, 64, 65 and 66 can be obtained quickly.

In the indicated value reading system of this embodiment, the image processing device 12 is connected to the monitor 16, and the image data of the camera 11 and the measuring instruments 51, 52, 53, 54, 55, and 56 processed by the image processing device 12 are read. Image data can be displayed on the monitor 16 . Therefore, the operator can confirm the visual angle of the camera 11 and the specified measuring instruments 51 , 52 , 53 , 54 , 55 , 56 .

In the indicated value reading system of the present embodiment, the image processing device 12 includes setting data necessary for processing, and indicated value data of the pointers 61, 62, 63, 64, 65, and 66 obtained by the image processing device 12. can be displayed on the monitor 16. Therefore, the operator can confirm the processing content and can easily confirm the processing error.

The indicated value reading method of the present embodiment includes a step of photographing the pointers 61, 62, 63, 64, 65, and 66 of the measuring instruments 51, 52, 53, 54, 55, and 56; , 63, 64, 65, 66 to obtain the indicated values of the pointers 61, 62, 63, 64, 65, 66; a step of deleting the image data and transmitting the obtained data of the indicated values of the pointers 61, 62, 63, 64, 65, 66; Therefore, the data of the indication values of the pointers 61, 62, 63, 64, 65, and 66 transmitted by the transmitting device 13 are numerical values as digital data, and thus are low-capacity data, and high-capacity data is transmitted at high speed. No equipment required. As a result, an increase in introduction cost can be suppressed, and the indicated value of the measuring instrument can be easily read.

The indicated value reading program of the present embodiment includes processing for photographing the pointers 61, 62, 63, 64, 65, and 66 of the measuring instruments 51, 52, 53, 54, 55, and 56; , 63, 64, 65, 66 to obtain the indicated values of the pointers 61, 62, 63, 64, 65, 66; The image processing apparatus (computer) 12 is made to execute the process of deleting the image data and transmitting the obtained data of the indication values of the pointers 61, 62, 63, 64, 65, and 66. FIG. Therefore, the data of the indication values of the pointers 61, 62, 63, 64, 65, and 66 transmitted by the transmitting device 13 are numerical values as digital data, and thus are low-capacity data, and high-capacity data is transmitted at high speed. No equipment required. As a result, an increase in introduction cost can be suppressed, and the indicated value of the measuring instrument can be easily read.

In the above-described embodiment, the image processing device 12 processes the image data of the pointers 61, 62, 63, 64, 65, and 66 captured by the camera 11 at predetermined intervals to obtain the indicated values. is configured to transmit the data of the indicated value, but it is not limited to this configuration. For example, when an operator inputs an operation command to the image processing device 12, the indicated value reading system 10 may be configured to operate.

In the above-described embodiment, the six measuring instruments 51, 52, 53, 54, 55, and 56 are configured to obtain the indicated values of the pointers 61, 62, 63, 64, 65, and 66. may be any number. Also, the measuring instruments 51, 52, 53, 54, 55, 56 are circular, and the indicator needles 61, 62, 63, 64, 65, 66 are rotatably supported at the center position. is not limited to For example, the measuring instrument may be rectangular, and the pointer may be linearly or arcuately slidably supported.

REFERENCE SIGNS LIST 10 indicated value reading system 11 camera 12 image processing device 12a image processing unit 12b image recognition unit 12c display creation unit 13 transmission device 14 storage unit 15 operation unit 16 monitor 17 power supply unit 20 monitoring system 21 control device 22 reception device 23 monitor 31 th 1 display section 32 2nd display section 33 3rd display section 34 4th display section 51, 52, 53, 54, 55, 56 measuring instrument 61, 62, 63, 64, 65, 66 pointer 71, 72, 73, 74,75,76 Scale 100 Field 200 Control Center

Claims (18)

a camera capable of photographing the pointer of the measuring instrument;
an image processing device connected to the camera and processing image data of the pointer photographed by the camera to obtain an indicated value of the pointer;
a transmission device for transmitting the data of the indication value of the pointer obtained by the image processing device;
with
A storage unit is provided for storing the indicated value of the pointer obtained by the image processing device,
The image processing device obtains the likelihood of the indication value of the pointer obtained by processing the image data of the pointer, and if the likelihood is lower than a preset threshold, the image data of the pointer in the storage unit;
An indicated value reading system characterized by: The image processing device obtains an indicated value of the pointer in each predetermined processing cycle, the transmission device transmits data of the indicated value of the pointer in each predetermined transmission cycle, and the likelihood is 2. The indicated value reading system according to claim 1, wherein, when the value is lower than the threshold value, the transmitting device transmits data of the indicated value of the pointer obtained last time by the image processing device. a camera capable of photographing the pointer of the measuring instrument;
an image processing device connected to the camera and processing image data of the pointer photographed by the camera to obtain an indicated value of the pointer;
a transmission device for transmitting the data of the indication value of the pointer obtained by the image processing device;
with
The image processing device obtains the likelihood of the indication value of the pointer obtained by processing the image data of the pointer, and obtains the indication value of the pointer for each predetermined processing cycle,
The transmitting device transmits the data of the indication value of the pointer in each predetermined transmission cycle, and when the likelihood is lower than a preset threshold value, the transmitting device transmits data indicating that the image processing device Transmitting the obtained data of the indicated value of the pointer;
An indicated value reading system characterized by: a camera capable of photographing the pointer of the measuring instrument;
an image processing device connected to the camera and processing image data of the pointer photographed by the camera to obtain an indicated value of the pointer;
a transmission device for transmitting the data of the indication value of the pointer obtained by the image processing device;
with
The image processing device obtains the likelihood of the indication value of the pointer obtained by processing the image data of the pointer, and determines that the likelihood is equal to or greater than a preset first threshold value of the pointer. Data of an indication value is transmitted by the transmission device, and if the likelihood is lower than the first threshold value and equal to or greater than a second threshold value smaller than the first threshold value, the indication value of the pointer is regarded as a similarity value for image matching. or corrected by the previous value and transmitted by the transmitting device;
An indicated value reading system characterized by: In the image processing device, the likelihood is lower than the first threshold and equal to or higher than the second threshold, and the absolute value of the difference between the indicated value of the pointer and the similarity value is equal to or lower than a preset deviation threshold. 5. The indicated value reading system according to claim 4 , wherein if there is, the indicated value of the pointer is corrected by the similar value and transmitted by the transmitting device. In the image processing device, the absolute value of the difference between the indicated value of the pointer and the similarity value is higher than the deviation threshold, the degree of similarity of the similarity value is equal to or higher than a third threshold, and the indicated value of the pointer and When the absolute value of the difference from the similarity value is less than or equal to the absolute value of the difference between the previous value and the similarity value, the indicated value of the pointer is corrected by the similarity value and transmitted by the transmitting device. 6. The indicated value reading system according to claim 5 . In the image processing device, the absolute value of the difference between the indicated value of the pointer and the similarity value is higher than the deviation threshold, the degree of similarity of the similarity value is equal to or higher than a third threshold, and the indicated value of the pointer and When the absolute value of the difference from the similarity value is higher than the absolute value of the difference between the previous value and the similarity value, the indicated value of the pointer is corrected by the previous value and transmitted by the transmitting device. 7. The indicated value reading system according to claim 5 or 6 . The indicated value reading system according to any one of claims 1 to 7, wherein the camera and the image processing device are connected by wire. 9. The image processing device according to any one of claims 1 to 8, wherein after processing the image data of the pointer to obtain the indicated value of the pointer, the image data of the pointer is deleted. The indicated value reading system according to item 1 . The operating range and unit of the pointer are assigned to the measuring device, and the image processing device processes the image data of the pointer to obtain a numerical value as the indicated value of the pointer. The indicated value reading system according to any one of claims 1 to 9 . The camera is capable of photographing a plurality of pointers of the measuring instrument, and the image processing device sequentially processes the image data of the plurality of pointers photographed by the camera to obtain an indicated value of the pointer. The indicated value reading system according to any one of claims 1 to 10, characterized in that: The image processing device processes image data captured by the camera to create image data of the measuring device, and obtains an indicated value of the pointer in the measuring device by deep learning of the image data of the measuring device. The indicated value reading system according to any one of claims 1 to 11, characterized in that: 2. A monitor is connectable to said image processing device, and image data of said camera and image data of said measuring instrument processed by said image processing device can be displayed on said monitor. 13. An indication reading system according to any one of claims 12 to 12. 14. The image processing device according to claim 13 , wherein the image processing device is capable of displaying on the monitor setting data necessary for processing and data of the indication value of the pointer obtained by the image processing device. Indicated value reading system. a step of photographing the pointer of the measuring instrument;
a step of processing the photographed image data of the pointer to obtain an indicated value of the pointer;
a step of deleting the image data of the pointer and transmitting the obtained data of the indicated value of the pointer;
has
determining the likelihood of the indicated value of the pointer obtained by processing the image data of the pointer, and storing the image data of the pointer when the likelihood is lower than a preset threshold;
An indicated value reading method characterized by: a step of photographing the pointer of the measuring instrument;
a step of processing the photographed image data of the pointer to obtain an indicated value of the pointer;
a step of deleting the image data of the pointer and transmitting the obtained data of the indicated value of the pointer;
has
The likelihood of the indication value of the pointer obtained by processing the image data of the pointer is obtained, and if the likelihood is equal to or greater than a preset first threshold value, the data of the indication value of the pointer is transmitted. On the other hand, when the likelihood is lower than the first threshold and equal to or greater than a second threshold smaller than the first threshold, the indication value of the pointer is corrected by the similarity value or the previous value of image matching and transmitted.
An indicated value reading method characterized by: a process of photographing the pointer of the measuring instrument;
a process of processing the photographed image data of the pointer to obtain an indicated value of the pointer;
a process of deleting the image data of the pointer and transmitting the obtained data of the indicated value of the pointer;
A process of obtaining the likelihood of the indication value of the pointer obtained by processing the image data of the pointer, and storing the image data of the pointer when the likelihood is lower than a preset threshold;
An indicated value reading program characterized by causing a computer to execute a process of photographing the pointer of the measuring instrument;
a process of processing the photographed image data of the pointer to obtain an indicated value of the pointer;
a process of deleting the image data of the pointer and transmitting the obtained data of the indicated value of the pointer;
The likelihood of the indication value of the pointer obtained by processing the image data of the pointer is obtained, and if the likelihood is equal to or greater than a preset first threshold value, the data of the indication value of the pointer is transmitted. On the other hand, when the likelihood is lower than the first threshold and equal to or greater than a second threshold smaller than the first threshold, a process of correcting the indication value of the pointer by the similarity value or the previous value of image collation and transmitting it. When,
An indicated value reading program characterized by causing a computer to execute

Images