A method for on-site video acquisition

Abstract

The present invention discloses a method for on-site video acquisition, belonging to the fields of intelligent construction sites, intelligent cities, intelligent factories, etc. The method includes obtaining the image gray-scale change rate of the image data transmitted and collected by a camera, and the sound intensity change rate of the sound data transmitted and collected by the camera; according to the image gray-scale change rate and the sound intensity change rate, combining with the on-site equipment failure status signal, querying a preset expert knowledge base, and inferring to obtain the video remote transmission correction coefficient of the camera; obtaining the current moment video data remote transmission bit rate and the frame number of the camera from the video remote transmission correction coefficient of the camera. By means of the present invention, on the premise of ensuring the requirements of the video system, the video data transmission and processing amount are reduced, the investment can be effectively saved, the system complexity can be reduced, and the system cost performance can be improved.

Description

Technical Field

[0001] The present invention belongs to the fields of smart construction sites, smart cities, smart factories, etc., and more specifically, relates to a method for collecting on-site video. Background Art

[0002] With the continuous intelligent upgrade of fields such as construction sites, factories, and cities, the number of cameras set in various scenarios in these fields has increased significantly, and the amount of video data generated by the cameras is huge. In order to ensure the reliability and real-time nature of this video data, the performance requirements for the corresponding video data transmission and processing facilities have also increased significantly.

[0003] In practical applications, the investment in video data transmission and processing facilities increases even faster with the improvement of their performance. Therefore, an optimized on-site video collection method needs to be adopted to reduce the amount of video data transmission and processing on the premise of meeting the requirements of the video system, which can effectively save investment, reduce system complexity, and improve system cost performance. Summary of the Invention

[0004] Aiming at the above defects or improvement requirements of the prior art, the present invention proposes a method for collecting on-site video, which can reduce the amount of video data transmission and processing on the premise of meeting the requirements of the video system, effectively save investment, reduce system complexity, and improve system cost performance.

[0005] To achieve the above object, the present invention provides a method for collecting on-site video, including:

[0006] Obtaining the image gray-scale change rate of the image data transmitted and collected by the camera, and the sound intensity change rate of the sound data transmitted and collected by the camera;

[0007] According to the image gray-scale change rate and the sound intensity change rate, combined with the on-site equipment failure status signal, query the preset expert knowledge base, and infer to obtain the video remote transmission correction coefficient of the camera;

[0008] Obtain the video data remote transmission bit rate and the number of frames of the camera at the current moment from the video remote transmission correction coefficient of the camera.

[0009] In some optional implementation schemes, from Obtain the image gray-scale change rate P% of the image data transmitted and collected by the camera, and the sound intensity change rate v% of the sound data transmitted and collected by the camera, v(n) represents the sound intensity value collected by the camera last time, v(n + 1) represents the sound intensity value collected by the camera this time, Pij(n) represents the gray-scale value of the middle pixel of the ij partition of the image collected by the camera last time, Pij(n + 1) represents the gray-scale value of the middle pixel of the ij partition of the image collected by the camera this time, i and j are image partition serial numbers, and l and m are integers.

[0010] In some alternative embodiments, the preset expert knowledge base includes a fact base and a rule base. The fact base contains various facts classified according to different system times and the sum of the image gray-scale change rate and the sound intensity change rate. The rule base is composed of rules established by combining various facts in the created fact base, and the premise facts in the rule base are matched with facts of different system times and different sums of the image gray-scale change rate and the sound intensity change rate.

[0011] In some alternative embodiments, according to the image gray-scale change rate and the sound intensity change rate, in combination with the on-site device fault status signal, query the preset expert knowledge base and infer the video remote transmission correction coefficient of the camera, including:

[0012] Match the known conditions input by the user with the facts in the fact base one by one and generate facts;

[0013] For any rule in the rule base, extract the premise facts of the rule, verify whether the extracted premise facts match the generated facts. If the extracted premise facts match the generated facts, it means that the rule is successfully matched; if the extracted premise facts do not match the generated facts, take the next rule for matching;

[0014] Output the conclusion of the successfully matched rule to obtain the video remote transmission correction coefficient K.

[0015] In some alternative embodiments, from obtain the current video data remote transmission bit rate M(t) and the frame rate Z(t) of the camera, where t represents the time at the location of the video system, K represents the video remote transmission correction coefficient of the camera, M0 represents the maximum video data remote transmission bit rate of the camera, and Z0 represents the maximum frame rate of the camera.

[0016] Generally speaking, compared with the prior art through the above technical solutions conceived by the present invention, the following beneficial effects can be achieved: By analyzing the knowledge base of the video image data, sound data, device status data, and time collected on-site, dynamically adjust the video data remote transmission bit rate and frame rate of the camera. When there are animal activities, device failures, etc. on-site, increase the video data remote transmission bit rate and frame rate of the camera; in other normal states, reduce the video data remote transmission bit rate and frame rate of the camera to achieve the purpose of saving video data transmission and processing volume. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic diagram of a method for on-site video acquisition provided by an embodiment of the present invention;

[0018] Figure 2 is a schematic diagram of the principle of an expert system for video remote transmission correction coefficient provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0020] As Figure 1 shown is a schematic diagram of a method for collecting on-site video provided by an embodiment of the present invention. The camera monitors on-site equipment in real time and transmits the collected image data and sound data to the video system, and obtains the image gray-scale change rate and the sound intensity change rate according to formula (1).

[0021]

[0022] Figure 1 In the formula, Q represents the on-site equipment failure status value. When Q = 1, it indicates that there is equipment failure; t represents the time at the location of the video system; v represents the sound intensity collected by the camera, with the unit of dB; P represents the image gray-scale value collected by the camera, and the value range is 0-255; M(t) represents the video data remote transmission bit rate of the camera at time t, with the unit of Mbps; Z(t) represents the number of frames of the camera at time t, with the unit of FPS.

[0023] In formula (1), v% represents the sound intensity change rate, with the unit of %; P% represents the image gray-scale change rate, with the unit of %; t represents the time at the location of the video system; v(n) represents the sound intensity value collected by the camera last time, with the unit of dB; v(n + 1) represents the sound intensity value collected by the camera this time, with the unit of dB; Pij(n) represents the gray-scale value of the middle pixel of the ijth partition of the image collected by the camera last time; Pij(n + 1) represents the gray-scale value of the middle pixel of the ijth partition of the image collected by the camera this time; i and j are the image partition serial numbers. In this embodiment, the integer value range is 0 to 2, indicating that the collected image is evenly divided into 9 regions in 3 rows and 3 columns.

[0024] According to the image gray-scale change rate and the sound intensity change rate, combined with the on-site equipment failure status signal Q, query the preset expert knowledge base that has been created, infer the video remote transmission correction coefficient of the camera, and then calculate the video data remote transmission bit rate and the number of frames according to formula (2).

[0025]

[0026] In Equation (2), M(t) represents the remote transmission bit rate of the video data of the camera at time t, with the unit of Mbps; Z(t) represents the number of frames of the camera at time t, with the unit of FPS; M0 represents the maximum remote transmission bit rate of the video data of the camera, with the unit of Mbps; Z0 represents the maximum number of frames of the camera, with the unit of FPS; K represents the video remote transmission correction coefficient of the camera, which is obtained by the inference of the expert system.

[0027] The principle of the expert system for the video remote transmission correction coefficient is as Figure 2 shown. The expert system for the video remote transmission correction coefficient is an inference system for the video remote transmission correction coefficient constructed based on a preset expert knowledge base. Figure 2 Among them, X represents the input of the expert system interface, including the system time, the fault status of on-site equipment, and the management input to the knowledge base; Y represents the output of the expert system, which here refers to the video remote transmission correction coefficient K; the video system interface refers to the video system computer; knowledge management is the knowledge maintenance such as adding, deleting, and modifying the knowledge in the knowledge base; the knowledge base is the collection of decision-making knowledge and experience knowledge of the decision-making experts for the video remote transmission correction coefficient; the inference engine is a set of programs that process the knowledge base according to the system time, the fault status of on-site equipment, the change rate of image gray scale, and the change rate of sound intensity, and feedback the inference result to the video system interface.

[0028] Among them, (1) Create a preset expert knowledge base;

[0029] (1.1) Knowledge collection

[0030] The expert knowledge of the video remote transmission correction coefficient is collected from the data summarized from practical experience.

[0031] When a fault occurs in the on-site equipment (Q = 1), according to the actual situation on-site, the common video remote transmission correction coefficient K = 100% is listed.

[0032] When no fault occurs in the on-site equipment (Q = 0), according to the actual situation on-site, a common video remote transmission correction coefficient table is listed, as shown in Table 1.

[0033] Table 1 Video remote transmission correction coefficient (%)

[0034]

[0035] (1.2) Knowledge expression

[0036] The rule method is used to express the expert knowledge of the video remote transmission correction coefficient. Its standard program structure is "if - then" (IF - THEN), that is, to evaluate a situation. If the situation is true, then take action. After the knowledge expression of the expert knowledge in Table 1 by the rule method, two parts, namely the fact base and the rule base, are generated.

[0037] 1) Generate the fact base

[0038] Divide and create facts according to the principles of different system times, different sums of the image gray-scale change rate and the sound intensity change rate. When the control requirements are increased, the facts are refined; when the control requirements are decreased, the facts are coarsened. Now, divide the facts according to the information in Table 1, and the established fact library is shown in Table 2, which contains facts such as "Fact 1",..., "Fact C".

[0039] Table 2 Fact Library

[0040]

[0041]

[0042] 2) Generate the rule library

[0043] Combine the facts in the created fact library to establish the rule library, as shown in Table 3, which contains facts such as "Rule 1A",..., "Rule 2C". Among them, the rule "Rule 1A" expresses the expert knowledge that "when the on-site equipment has no fault (Q = 0), if the time is during the day AND (P% + v%) < 10%; then the video remote transmission correction coefficient K = 10%".

[0044] Table 3 Rule Library

[0045] Serial number Rule Rule 1A IF Fact 1 AND Fact A; THEN K = 10% Rule 1B IF Fact 1 AND Fact B; THEN K = 50% Rule 1C IF Fact 1 AND Fact B; THEN K = 100% Rule 1A IF Fact 1 AND Fact A; THEN K = 15% Rule 1B IF Fact 1 AND Fact B; THEN K = 60% Rule 1C IF Fact 1 AND Fact B; THEN K = 100%

[0046] (2) Video Remote Transmission Correction Coefficient Knowledge Reasoning

[0047] The video remote transmission correction coefficient expert system performs knowledge reasoning through an inference engine to obtain the video remote transmission correction coefficient of the camera in various situations.

[0048] (2.1) Reasoning Method

[0049] The inference engine of the video remote transmission correction coefficient expert system adopts the forward reasoning method. It processes the facts and rules in the system knowledge base for known conditions such as system time, image gray-scale change rate, and sound intensity change rate. Its reasoning principle is:

[0050] If fact M is true and there is a rule "TF M THEN N", then N is true.

[0051] Therefore, if the known conditions input by the user satisfy Fact 1 and Fact A in the fact library, and there is a rule "IF Fact 1 AND Fact A; THEN K = 10%" in the rule library; then the video remote transmission correction coefficient K of the camera can be obtained as K = 10%.

[0052] The working process of the inference engine is as follows:

[0053] 1) Match the known conditions input by the user with the facts in the fact library one by one and generate facts;

[0054] 2) The rule premises in the rule base are matched with the generated system time, the sum of the image gray-scale change rate and the sound intensity change rate facts; take out the <premises> of each rule and verify whether the extracted premise facts match the generated facts. If the extracted premise facts match the generated facts, it means that the rule is successfully matched; if the extracted premise facts do not match the generated facts, then take the next rule for matching.

[0055] 3) Output the <conclusions> of the rules that match successfully to obtain the video remote transmission correction coefficient K.

[0056] Combined with the specifications of the actual camera, substitute into Equation (2) to obtain the video data remote transmission bit rate and frame rate of this camera.

[0057] It should be noted that according to the implementation needs, each step / component described in this application can be split into more steps / components, or two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0058] Those skilled in the art can easily understand that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for on-site video acquisition, characterized in that, Including: Obtaining the image gray change rate of the image data collected by the camera transmission, and the sound intensity change rate of the sound data collected by the camera transmission; According to the image gray change rate and the sound intensity change rate, combined with the on-site equipment failure status signal, query the preset expert knowledge base, and infer the video remote transmission correction coefficient of the camera. The preset expert knowledge base includes a fact base and a rule base. The fact base is various facts divided according to the principles of different system times, different sums of the image gray change rate and the sound intensity change rate. The rule base is a rule established by combining various facts in the created fact base, and matches the premise facts in the rule base with the facts of different system times, different sums of the image gray change rate and the sound intensity change rate; Obtaining the video data remote transmission bit rate and the frame number of the camera at the current moment from the video remote transmission correction coefficient of the camera; According to the image gray change rate and the sound intensity change rate, combined with the on-site equipment failure status signal, query the preset expert knowledge base, and infer the video remote transmission correction coefficient of the camera, including: Matching the known conditions input by the user with the facts in the fact base item by item and generating facts; For any rule in the rule base, extracting the premise facts of the rule, verifying whether the extracted premise facts match the generated facts. If the extracted premise facts match the generated facts, it means that the rule match is successful; if the extracted premise facts do not match the generated facts, take the next rule for matching; Outputting the conclusion of the rule with successful match to obtain the video remote transmission correction coefficient K.

2. The method according to claim 1, wherein From Obtain the image gray-scale change rate P% of the image data transmitted and collected by the camera, and the sound intensity change rate v% of the sound data transmitted and collected by the camera. v(n) represents the sound intensity value collected by the camera last time, v(n + 1) represents the sound intensity value collected by the camera this time, Pij(n) represents the gray-scale value of the middle pixel of the ij-th partition of the image collected by the camera last time, and Pij(n + 1) represents the gray-scale value of the middle pixel of the ij-th partition of the image collected by the camera this time. i and j are the image partition numbers.

3. The method according to claim 2, wherein From obtain the remote transmission bit rate M(t) of the video data at the current moment and the number of frames Z(t) of the camera, where t represents the time at the location of the video system, K represents the video remote transmission correction coefficient of the camera, M0 represents the maximum remote transmission bit rate of the video data of the camera, and Z0 represents the maximum number of frames of the camera.

Images