Data acquisition system and data acquisition method

Abstract

The invention provides a data acquisition system and a data acquisition method. The data acquisition system comprises a photographing device, a storage device, a host device and a server. The host device is coupled to the photographing device and the storage device, and is used for obtaining the monitoring image through the photographing device and backing up the monitoring image to the storage device. The server is coupled to the host device and the storage device, and is used for reading the monitoring image from the host device or the storage device, and obtaining the monitoring data from the monitoring image according to a preset identification frame. The server executes the target detection module to identify a numerical value in the monitoring data, and the server determines whether the confidence of the numerical value is greater than or equal to a first confidence threshold. When the server judges that the confidence coefficient of the numerical value is greater than or equal to a first confidence coefficient threshold value, the server judges whether the numerical value is within a safety range or not; and when the server judges that the numerical value is not in the safety range, the server generates an abnormal notification. The data acquisition system and the data acquisition method can provide a data acquisition function.

Description

Technical Field

[0001] The invention relates to a system, in particular to a data acquisition system and a data acquisition method. Background Art

[0002] The conventional data collection method is to record the equipment data manually, but this data collection method is not only inefficient, but also the credibility of the recorded data is not high. On the other hand, if the existing equipment is upgraded in batches to achieve the automatic equipment data transmission function, it will not only increase the cost of equipment transformation, but also affect the time of application deployment. Summary of the invention

[0003] The present invention provides a data acquisition system and a data acquisition method, which can perform real-time data acquisition by means of camera deployment and combining with improving existing artificial intelligence technology.

[0004] The data acquisition system of the present invention includes a photographic device, a storage device, a host device and a server. The host device is coupled to the photographic device and the storage device, and obtains monitoring images through the photographic device, and backs up the monitoring images to the storage device. The server is coupled to the host device and the storage device. The server is used to read monitoring images from the host device or the storage device, and obtain monitoring data from the monitoring images according to a default identification box. The server executes a target detection module to identify a numerical value in the monitoring data, and determines whether the confidence of the numerical value is greater than or equal to a first confidence threshold. When the server determines that the confidence of the numerical value is greater than or equal to the first confidence threshold, the server determines whether the numerical value is within a safe range. When the server determines that the numerical value is not within a safe range, the server generates an abnormal notification.

[0005] The data acquisition method of the present invention includes the following steps: obtaining a monitoring image through a photographic device; backing up the monitoring image to a storage device; reading the monitoring image and obtaining monitoring data from the monitoring image according to a preset identification frame; executing a target detection module to identify a numerical value in the monitoring data; determining whether the confidence of the numerical value is greater than or equal to a first confidence threshold; when the confidence of the numerical value is greater than or equal to the first confidence threshold, determining whether the numerical value is within a safe range; and generating an abnormal notification when the numerical value is not within the safe range.

[0006] Based on the above, the data acquisition system and data acquisition method of the present invention can obtain monitoring data through a photographic device, and can automatically judge the monitoring data to determine whether to generate an early warning notification. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The accompanying drawings are included to provide a further understanding of the present invention and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments of the present invention and together with the description serve to explain the principles of the present invention.

[0008] Figure 1 is a schematic diagram of a data acquisition system according to an embodiment of the present invention;

[0009] Figure 2 is a flow chart of confidence judgment of a numerical value according to an embodiment of the present invention;

[0010] Figure 3 is a flow chart of setting an identification frame according to an embodiment of the present invention;

[0011] Figure 4 is a schematic diagram of a surveillance image according to an embodiment of the present invention;

[0012] Figure 5 is a schematic diagram of an application program interface according to an embodiment of the present invention;

[0013] Figure 6 is a flow chart of identifying a numerical value according to an embodiment of the present invention;

[0014] Figure 7 is a schematic diagram of a numerical profile of an embodiment of the present invention;

[0015] Figure 8 is a flow chart of adjusting the identification frame according to an embodiment of the present invention;

[0016] Fig. 9 is a schematic diagram of adjusting a preset identification frame according to an embodiment of the present invention;

[0017] Fig.10 is a schematic diagram of an application program interface according to an embodiment of the present invention.

[0018] Description of Reference Numerals

[0019] 100: data acquisition system;

[0020] 110: host device;

[0021] 120: Photographic device;

[0022] 130: server;

[0023] 140: storage device;

[0024] 200: terminal device;

[0025] 400, 900: surveillance images;

[0026] 401-404, 911-913, 921-923: identification frame;

[0027] 500: application program interface;

[0028] 701-705: numerical profile;

[0029] S210~S280, S310~S370, S601~S611, S810~S830: steps. DETAILED DESCRIPTION

[0030] Reference will now be made in detail to exemplary embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Whenever possible, the same reference numerals are used in the drawings and the description to refer to the same or like parts.

[0031] Figure 1 is a schematic diagram of a data acquisition system 100 according to an embodiment of the present invention. Figure 1 In this embodiment, the data acquisition system 100 includes a host device 110, a photographic device 120, a server 130, and a storage device 140. The host device 110 is coupled to the photographic device 120, the server 130, and the storage device 140. The server 130 is coupled to the storage device 140. In this embodiment, the photographic device 120 may be set at the local end, and the host device 110, the server 130, and the storage device 140 may be set in the cloud. The host device 110 may be connected to the photographic device 120, the server 130, and the storage device 140 via a wired or wireless communication method. The server 130 may be connected to the storage device 140 via a wired or wireless communication method. In one embodiment, the data acquisition system 100 may include multiple photographic devices.

[0032] In the present embodiment, the host device 110 may be a network video recorder (NVR). The photographic device 120 may include a camera, and transmits the image data to the host device 110, for example, via a universal serial bus (USB), Bluetooth, or Wi-Fi, but the present invention is not limited thereto. In the present embodiment, the photographic device 120 may be used to photograph a device having a display capable of displaying a numerical display screen, so as to photograph the numerical display screen of the device. And the image data of the numerical display screen is transmitted to the server 130 and / or the storage device 140. In the present embodiment, the server 130 may be a cloud server, and may include a processor and a memory. In the present embodiment, the storage device 140 may be a network attached storage (NAS) device, and may be used to back up or temporarily store the image data provided by the photographic device 120 and the related data generated by the host device 110.

[0033] In this embodiment, the terminal device 150 may be, for example, a desktop computer, a laptop computer, a smart phone, a tablet computer or other computing devices. In this embodiment, the user may connect to the server 130 and / or the storage device 140 through the terminal device 150. The server 130 and / or the storage device 140 may respectively provide an application programming interface (API) for the terminal device 150 to connect and execute the image monitoring functions described in the various embodiments of the present invention.

[0034] Figure 2 FIG. 1 is a flowchart of determining the confidence level of a numerical value according to an embodiment of the present invention. Figure 1 as well as Figure 2 , the data acquisition system 100 may execute the following steps S210 to S270. In step S210, the host device 110 obtains the surveillance image through the camera device. In step S220, the host device 110 backs up the surveillance image to the storage device 140. In step S230, the server 130 reads the surveillance image and obtains the surveillance data from the surveillance image according to the preset identification frame. In this embodiment, the user may operate the terminal device 200 to connect to the server 130 and execute the corresponding application program interface to analyze the surveillance image to obtain the surveillance data.

[0035] In step S240, the server 130 executes a target detection module to identify the numerical value in the monitoring data. In the present embodiment, the server 130 may store a target detection module, and the target detection module may include a target detection method, such as yolo5, but the present invention is not limited thereto. In one embodiment, the target detection module may also include a neural network model or a deep learning model for image recognition, such as a convolutional neural network (CNN), a YOLO (You Only Look Once) model, or a similar neural network.

[0036] In step S250, the server 130 determines whether the confidence of the numerical value is greater than or equal to a first confidence threshold. In step S260, when the confidence of the numerical value is greater than or equal to the first confidence threshold, the server 130 determines whether the numerical value is within a safe range. The first confidence threshold may be 60%, for example. In step S270, when the numerical value is not within a safe range, the server 130 generates an abnormality notification. Therefore, the data acquisition system 100 and the data acquisition method of the present embodiment can realize a remote device monitoring function. The data acquisition system 100 and the data acquisition method of the present embodiment can monitor the current numerical value of the device in real time to automatically determine whether the current numerical value of the device is abnormal, and decide whether to generate an abnormality notification.

[0037] Figure 3 is a flow chart of setting an identification frame according to an embodiment of the present invention. Figure 4 FIG. 1 is a schematic diagram of an initial monitoring image according to an embodiment of the present invention. Figure 1 , Figure 3 as well as Figure 4 In this embodiment, the user can connect to the server 130 through the terminal device 200 to read, for example, Figure 4 The data acquisition system 100 may perform the following steps S310 to S370. In step S310, the server 130 reads the initial surveillance image 400 from the host device 110 or the storage device 140. In step S320, the server 130 identifies the numerical image area in the initial surveillance image 400. In step S330, the server 130 generates, for example, the identification boxes 401 and 402 to be bound. In step S340, the server 130 may determine whether to add a new identification area. The identification area may be added manually by the user or by the server 130 in the form of image recognition. If so, the server 130 executes step S330 again to add the identification boxes 403 and 404 to be bound. If not, in step S350, the server 130 binds the identification boxes 401 to 404 to be bound with the corresponding acquisition devices. In this embodiment, when the server 130 sets a new identification frame to be bound according to the control instruction, the server 130 binds the new identification frame to be bound with the corresponding device.

[0038] In step S360, the server 130 determines whether all the numerical image areas have been bound. If not, the server 130 executes step S340 again. If so, in step S370, the server 130 completes the setting of the preset identification frame. In this embodiment, the photographic device 120 can shoot the acquisition device and obtain the initial monitoring image 400, and the server 130 can perform image recognition on the initial monitoring image 400 and detect multiple numerical image areas. The server 130 generates multiple identification frames near the multiple numerical image areas, and each identification frame corresponds to different device parameters. In this embodiment, when all the identification frames are bound, the server 130 only needs to perform image recognition on the image corresponding to the position of the preset identification frame in the subsequent monitoring image to obtain multiple real-time device parameters.

[0039] Figure 5 is a schematic diagram of an application program interface according to an embodiment of the present invention, wherein the application program interface can be implemented as follows Figure 1 The data acquisition system 100 shown generates Figure 1 , Figure 4 as well as Figure 5 , the server 130 may display Figure 5The application program interface 500 (e.g., a web page) is shown. In this embodiment, after the server 130 detects the numerical image area in the surveillance image 400 and generates a preset identification frame according to the above process, the server 130 can obtain monitoring data from the surveillance image according to the preset identification frame, and display the obtained monitoring data on the application program interface 500. In this embodiment, the photographing device 120 can, for example, obtain multiple surveillance images, so that the server 130 can obtain multiple monitoring data from the multiple surveillance images and present them in sequence in the data list of the application program interface 500. The user can obtain the monitoring data of all acquisition devices by accessing the application program interface 500 from the terminal device 150.

[0040] In the application program interface 500, the application program interface 500 may display information such as the camera number, date and device number, and may include a function option for adding a new acquisition device. In addition, the device information corresponding to a specific device number may also include information and function options such as data name, data range, implementation value, decimal place, box selection data and data error.

[0041] Figure 6 is a flow chart of identifying a numerical value according to an embodiment of the present invention. Figure 7 is a schematic diagram of the numerical profile of an embodiment of the present invention. Figure 1 , Figure 6 and Figure 7 In this embodiment, the data acquisition system 100 may execute the following steps S601 to S611 to make a confidence judgment on the numerical value. In step S601, the server 130 may execute the target detection module to identify the numerical value in the surveillance image. In step S602, the server 130 determines the confidence of the numerical value. In this embodiment, when the server 130 determines that the confidence of the numerical value is greater than or equal to the second confidence threshold (confidence ≥ 80%), in step S603, the server 130 first identifies the decimal point of the numerical value. In this embodiment, the server 130 performs image preprocessing on the numerical image corresponding to the preset identification box, and determines multiple numerical contours in the numerical image. The image preprocessing may include color filtering, grayscale, binarization, and related morphological processing. The server 130 determines that the one with the smallest contour area among multiple numerical contours is the decimal point. For example, if Figure 7 As shown, the server 130 can identify multiple numerical contours 701-705 in the numerical image 700 after image preprocessing, and calculate the contour area and the centroid. In addition, the server 130 can determine that the position corresponding to the contour with the smallest area is the decimal point. The server 130 can determine that the decimal point is located at the third position according to the order of the decimal points in the numerical contours 701-705. In this regard, the server 130 can determine that the integer part of the numerical value is 2 digits, and the decimal part is 2 digits (5-3=2).

[0042] Next, in step S604, the server 130 determines whether the value is within the safety range. The second confidence threshold is greater than the first confidence threshold. The second confidence threshold may be, for example, 80%, and the first confidence threshold may be, for example, 60%. In addition, when the server 130 determines that the confidence of the value is greater than or equal to the first confidence threshold, and less than the second confidence threshold (60% ≤ confidence < 80%), in step S606, the server 130 may first generate an early warning notification, and then execute in step S603. In step S604, the server 130 determines whether the value is within the safety range. If so, in step S605, the server 130 completes the identification of the value. If not, in step S607, the server 130 generates an abnormal notification.

[0043] When the server 130 determines that the confidence of the numerical value is less than the first confidence threshold (confidence <60%), in step S608, the server 130 executes the target detection module to perform mixed recognition on the numerical value. In step S609, the server 130 determines whether the numerical value is an abnormal code. If so, in step S610, the server 130 re-identifies whether the numerical value exceeds a preset number of times. If so, the server 130 determines that the numerical value is an abnormal code. In step S611, the server 130 generates an abnormal notification. If not, the server 130 determines that the numerical value is not an abnormal code. The server 130 re-executes step S601 to re-identify the numerical value. In addition, when the server 130 re-identifies the numerical value more than a preset number of times, the server 130 may generate an abnormal notification.

[0044] Figure 8 is a flow chart of adjusting an identification frame according to an embodiment of the present invention. Fig. 9 FIG. 1 is a schematic diagram of adjusting a preset identification frame according to an embodiment of the present invention. Figure 1 , Figure 8 and Fig. 9 In this embodiment, the data acquisition system 100 may execute the following steps S810 to S830 to adjust the preset identification frame. Fig. 9In the surveillance image 900 shown in FIG. 1 , when the server 130 cannot obtain surveillance data from the surveillance image 900 according to the preset identification frames 911 to 913, in step S810, the server 130 calculates a new range centroid of the numerical image in the surveillance image 900. In step S820, the server 130 compares the new range centroid with the old range centroid to calculate the adjustment angle. In step S830, the server 130 adjusts the preset identification frames 911 to 913 according to the adjustment angle. In this regard, the server 130 can obtain surveillance data from the surveillance image 900 again according to the adjusted preset identification frames 921 to 923. In addition, when the server 130 determines that the number of adjustments of the preset identification frame exceeds the adjustment number threshold, the server 130 generates another abnormal notification to remind the user.

[0045] Fig.10 FIG. 1 is a schematic diagram of an application program interface according to an embodiment of the present invention. Figure 1 as well as Fig.10 In this embodiment, different camera devices may have different camera numbers, and one camera device may shoot one or more devices and obtain multiple surveillance images. The server 130 may analyze multiple surveillance images to obtain multiple device data. In addition, the user may log in from the terminal device 200. Fig.10 The operation page of the application program interface 1000 of the server 130 is shown, and the operating status of multiple cameras and the status of data acquisition can be obtained from the application program interface 1000. For example, when a camera status shows a fault, the customer can know that the corresponding photographic device is online, but the host device 110 has not added any new recording content. Alternatively, when the camera status shows offline, it means that the camera is powered off or faulty, and personnel need to manually replace the camera so that the data acquisition system 100 can resume normal operation. In addition, the application program interface 1000 can also indicate the camera status by presenting different colors, but the present invention is not limited to this.

[0046] In summary, the data acquisition system and data acquisition method of the present invention can capture the equipment to obtain real-time monitoring images, and can also identify the values ​​in the monitoring images to obtain monitoring data. The data acquisition system and data acquisition method of the present invention can determine whether to generate an abnormal notification based on the confidence level of the value, whether the value is within a safe range, and other preset conditions.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data acquisition system, characterized in that: include: Photographic installations; Storage device; A host device coupled to the photographing device and the storage device, and used to obtain monitoring images through the photographing device and back up the monitoring images to the storage device; as well as A server is coupled to the host device and the storage device, and is used to read the monitoring image from the host device or the storage device, and obtain monitoring data from the monitoring image according to a preset identification frame. wherein the server executes a target detection module to identify a value in the monitoring data, and the server determines whether a confidence level of the value is greater than or equal to a first confidence threshold, When the server determines that the confidence of the value is greater than or equal to the first confidence threshold, the server determines whether the value is within a safe range. When the server determines that the value is not within a safe range, the server generates an exception notification.

2. The data acquisition system according to claim 1, characterized in that: The server reads an initial monitoring image from the host device or the storage device, and identifies a numerical image region in the initial monitoring image to generate an identification frame to be bound. The server binds the identification frame to be bound with the corresponding device to determine the preset identification frame.

3. The data acquisition system according to claim 2, characterized in that: When the server sets a new identification frame to be bound according to the control instruction, the server binds the new identification frame to be bound with the corresponding device.

4. The data acquisition system according to claim 1, characterized in that: When the server determines that the confidence of the value is less than the first confidence threshold, the server executes the target detection module to perform mixed recognition on the value, and determines whether the value is an abnormal code. When the server determines that the value is an abnormal code, the server generates the abnormal notification. When the server determines that the value is not an abnormal code, the server re-identifies the value.

5. The data acquisition system according to claim 4, characterized in that: When the server re-identifies the value for more than a preset number of times, the server generates the abnormal notification.

6. The data acquisition system according to claim 1, characterized in that: When the server determines that the confidence of the value is greater than or equal to a second confidence threshold, the server first identifies the decimal point of the value, and then determines whether the value is within the safety range. The second confidence threshold is greater than the first confidence threshold.

7. The data acquisition system according to claim 6, characterized in that: When the server determines that the confidence of the numerical value is greater than or equal to the first confidence threshold and less than the second confidence threshold, the server generates an early warning notification.

8. The data acquisition system according to claim 6, characterized in that: The server performs image preprocessing on the numerical image corresponding to the preset identification frame, and determines a plurality of numerical contours in the numerical image. The server determines that the one with the smallest contour area among the multiple numerical contours is the decimal point.

9. The data acquisition system according to claim 1, characterized in that: When the server cannot obtain monitoring data from the monitoring image according to the preset identification frame, the server calculates a new range centroid of the numerical image in the monitoring image, and the server compares the new range centroid with the old range centroid to calculate the adjustment angle. The server adjusts the preset identification frame according to the adjustment angle.

10. The data acquisition system according to claim 1, characterized in that: When the server determines that the number of adjustments to the preset identification frame exceeds an adjustment number threshold, the server generates another abnormal notification.

11. A data collection method, characterized in that: include: Obtain surveillance images through photographic devices; Backing up the surveillance images to a storage device; Reading the surveillance image, and obtaining surveillance data from the surveillance image according to a preset identification frame; Executing a target detection module to identify values ​​in the monitoring data; Determining whether the confidence level of the value is greater than or equal to a first confidence level threshold; When the confidence level of the value is greater than or equal to the first confidence level threshold, determining whether the value is within a safe range; as well as When the value is not within the safe range, an abnormality notification is generated.

12. The data collection method according to claim 11, characterized in that: Also includes: Read the initial surveillance image; Identifying a numerical image region in the initial monitoring image to generate an identification frame to be bound; as well as The identification frame to be bound is bound to a corresponding device to determine the preset identification frame.

13. The data collection method according to claim 12, characterized in that: Also includes: When a new identification frame to be bound is set according to the control instruction, the new identification frame to be bound is bound to the corresponding device.

14. The data collection method according to claim 11, characterized in that: Also includes: When the confidence of the value is less than the first confidence threshold, executing the target detection module to perform mixed recognition on the value, and determining whether the value is an abnormal code; When the value is an exception code, generating the exception notification; as well as When the value is not an abnormal code, the value is re-identified.

15. The data collection method according to claim 14, characterized in that: Also includes: When the value is re-identified more than a preset number of times, the abnormality notification is generated.

16. The data collection method according to claim 11, characterized in that: Also includes: When the confidence of the value is greater than or equal to a second confidence threshold, first identify the decimal point of the value, and then determine whether the value is within the safety range. The second confidence threshold is greater than the first confidence threshold.

17. The data collection method according to claim 16, characterized in that: Also includes: When the confidence of the numerical value is greater than or equal to the first confidence threshold and less than the second confidence threshold, an early warning notification is generated.

18. The data collection method according to claim 16, characterized in that: Also includes: Performing image preprocessing on the numerical image corresponding to the preset identification frame; as well as determining a plurality of numerical contours in the numerical image, Among the multiple numerical contours, the one with the smallest contour area is the decimal point.

19. The data collection method according to claim 11, characterized in that: Also includes: When the monitoring data cannot be obtained from the monitoring image according to the preset identification frame, calculating a new range centroid of the numerical image in the monitoring image; comparing the new range centroid with the old range centroid to calculate an adjustment angle; as well as The preset identification frame is adjusted according to the adjustment angle.

20. The data collection method according to claim 11, characterized in that: Also includes: When the number of times the preset identification frame is adjusted exceeds the adjustment number threshold, another abnormal notification is generated.

Images