A multi-source device display process monitoring system and method based on the Internet of Things

Abstract

The invention discloses a multi-source device display process supervision system and method based on the Internet of Things, designs the technical field of display device management, collects the historical records of a certain multi-source display device in displaying a target event, compiles them into a feature template, calculates the similarity between the sampled display screen and the feature template, samples the real-time display screen of each data source for several times, calculates the discrete values ​​of the similarity of the same data source respectively, and continues to sample when the discrete value of the similarity of the real-time display screen of the data source is greater than the discreteness evaluation threshold, calculates the weight value of the real-time display screen of each data source according to the number of sampling times corresponding to each data source, calculates the weight value of the real-time display screen of each data source when judging whether the target event occurs according to the weight value, uses the weight value as the voting coefficient corresponding to each data source, and each display screen votes on whether the target event occurs.

Description

Technical Field

[0001] The present invention relates to the technical field of display device management, and in particular to a multi-source device display process monitoring system and method based on the Internet of Things. Background Art

[0002] Multi-source display devices are a type of multi-source device that connects multi-channel signals to the same display device. They can simultaneously display images transmitted by multiple channels and are widely used in industrial testing or information transmission. By displaying multi-channel images at the same time, the management efficiency of relevant staff can be improved, such as obtaining information from multiple channels at the same time to judge the overall situation of an event.

[0003] However, due to the large number of display channels of multi-source display devices, relevant staff cannot pay attention to the images of all channels when monitoring the display of multi-source devices, so it is inevitable that information in the display images will be missed, which will affect the relevant staff's judgment on the overall situation of the incident. Summary of the invention

[0004] The purpose of the present invention is to provide a multi-source device display process supervision system and method based on the Internet of Things to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solution: a multi-source device display process supervision method based on the Internet of Things.

[0006] Step S100: collecting display images of various data sources of a multi-source display device, setting the display images as target events when they meet a certain feature, collecting historical records of the multi-source display device in displaying the target event, and compiling them into a feature template;

[0007] Step S200: sampling the real-time display pictures of each data source of the multi-source display device, calculating the similarity between the sampled display pictures and the feature template, sampling the real-time display pictures of each data source several times, and calculating the discrete values ​​of the similarity of the same data source respectively;

[0008] Step S300: setting a discreteness evaluation threshold, when the discreteness value of the real-time display screen similarity of the data source is greater than the discreteness evaluation threshold, continuing sampling, when the similarity between the newly added sampled display screen and the feature template makes the discreteness value of the similarity decrease, stopping further sampling;

[0009] Step S400: Calculate the weight value of the real-time display screen of each data source according to the sampling times corresponding to each data source, wherein the weight value is inversely proportional to the sampling times;

[0010] Step S500: Calculate the weight of the real-time display screen of each data source when judging whether the target event occurs according to the weight value, use the weight as the voting coefficient corresponding to each data source, and each display screen votes on whether the target event occurs to judge whether the target event occurs.

[0011] Furthermore, step S100 includes:

[0012] Step S101: at least two data sources are connected to the multi-source display device, wherein each data source corresponds to a picture display channel, and information collected from each data source is displayed in the corresponding picture display channel;

[0013] Step S102: acquiring the historical records displayed by each screen display channel when the target event occurs, and extracting features from the historical records displayed by each screen display channel to obtain feature screens corresponding to each screen display channel when the target event occurs;

[0014] Step S103: all picture display channels in the multi-source display device are arranged in order, and according to the arrangement order of the picture display channels, characteristic pictures corresponding to the picture display channels are collected to obtain characteristic templates corresponding to the target event;

[0015] When displaying data through a multi-source display device, the target event is related to the images displayed in each image display channel. Therefore, the characteristics of the target event are regarded as a combination of the characteristics of the images displayed in each image display channel. After obtaining the rules of the target event, the characteristics in each display image are extracted respectively, and then the characteristics are spliced ​​to obtain the characteristic template of the target event.

[0016] Further, step S200 includes:

[0017] Step S201: setting a unit sampling period with a time length of T0 and a unit sampling number a, sampling the pictures displayed in each picture display channel of the multi-source display device after entering any unit sampling period to obtain a sampled picture corresponding to each picture display channel, wherein the number of sampled pictures of each picture display channel is a;

[0018] Step S202: comparing the sampled images of each image display channel with the characteristic images of the corresponding image display channel to obtain the similarity between each sampled image and the characteristic image, and classifying and collecting the similarities according to the corresponding image display channel to obtain a first similarity data group corresponding to each image display channel;

[0019] Step S203: calculating the discrete values ​​corresponding to the data in the first similarity data group of each picture display channel, and matching the discrete values ​​with the picture display channels;

[0020] The discrete value is used to measure the degree of discreteness of similarity data. When the fluctuation range of similarity data is larger, the discreteness of data is larger and the discrete value is larger, which means that the degree of similarity between the picture displayed by the picture display channel and the feature template within the sampling range is more uncertain.

[0021] Furthermore, step S300 includes:

[0022] Step S301: Compare the discrete values ​​corresponding to each screen display channel with the discreteness evaluation threshold, obtain the screen display channels corresponding to the discrete values ​​greater than or equal to the discreteness evaluation threshold, and merge the screen display channels into a to-be-detected group, wherein the discrete value corresponding to the i-th screen display channel in the to-be-detected group is recorded as qi;

[0023] Step S302: for the picture display channel in the to-be-detected group, a number of sampled pictures are collected again, and similarities between the sampled pictures and corresponding feature pictures are calculated respectively, to obtain a second similarity data group corresponding to the picture display channel in the to-be-detected group;

[0024] Step S303: obtaining all data in the first similarity data group and the second similarity group of the i-th screen display channel in the group to be detected, calculating the discrete value of all the data and recording it as q'i, and when q'i≤qi, proceeding to step S400;

[0025] Step S304: when q'i>qi, the i-th picture display channel in the to-be-detected group is re-integrated into the to-be-detected group, and q'i is used as the new qi to repeat steps S301 to S302 until the output condition of step S303 is met and the loop is stopped;

[0026] When the discrete value of the similarity is too large, it means that the uncertainty of the similarity is high, and further sampling of the similarity is needed to determine the difference between the picture displayed in the picture display channel in the current state and the feature template. After adding several groups of sampling values, the discrete value becomes smaller, indicating that the degree of difference tends to be stable.

[0027] Furthermore, step S400 includes:

[0028] Step S401: Obtain the total number of sampling times corresponding to each picture display channel of the multi-source display device, wherein the total number of sampling times corresponding to the j-th picture display channel in the multi-source display device is recorded as NA j , the weight coefficient corresponding to the j-th screen display channel is recorded as μ j , μ j =a / NAj ;

[0029] Step S402: setting the initial weight value k0, the weight value corresponding to the j-th picture display channel in the multi-source display device is p j , p j =k0×μ j .

[0030] Furthermore, step S500 includes:

[0031] Step S501: obtaining r historical records of comparison between a sampled image and a corresponding feature image of any image display channel of the multi-source display device, wherein one historical record corresponds to one comparison between the sampled image and the corresponding feature image, and merging the similarities in the r historical records into a set R;

[0032] Step S502: Obtain the maximum similarity value e in R max and the minimum similarity e min , set the similarity reference range (c, d), where c < e min , d>e max ;

[0033] Step S503: Calculate the average similarity values ​​corresponding to the image display channels of the multi-source display device, wherein the average similarity value corresponding to the j-th image display channel in the multi-source display device is ave j , where h m represents the mth similarity of the jth picture display channel;

[0034] Step S504: Calculate the voting coefficient vote corresponding to the j-th picture display channel j ,

[0035] Step S505: Calculate the total number of votes V of all picture display channels in the multi-source display device. Among them, vote f represents the voting coefficient corresponding to the f-th picture display channel of the multi-source display device, and g represents the total number of picture display channels included in the multi-source display device;

[0036] Step S506: when V>0, it is determined that the target event has occurred, and when V≤0, it is determined that the target event has not occurred, and information reminders are given to relevant personnel according to the determination results;

[0037] The target event is judged whether it occurs by voting on each screen display channel. The closer the similarity between each sampled screen in the screen display channel and the feature template is, the more credible the similarity is. At this time, the voting power of the screen display channel is higher. The greater the degree of dispersion, the lower the credibility of the similarity is. The degree of dispersion of the similarity data is measured by the number of supplementary samples.

[0038] Set the similarity reference range (c, d). When the similarity is close to c, it means that the probability of the target event is low. When the similarity is close to d, it means that the probability of the target event is high. After all screen display channels vote, compare the total number of votes with the difference between the two boundaries to determine whether the target event occurs.

[0039] In order to better implement the above method, a multi-source device display process supervision system based on the Internet of Things is also proposed, and the system includes:

[0040] A feature template management module, a discrete value calculation module, a cyclic sampling module, a weight calculation module and a voting judgment module, wherein the feature template management module is used to manage the feature template of the target event, the discrete value calculation module is used to calculate the discrete value of the similarity, the cyclic sampling module is used to perform cyclic control on the picture sampling, the weight calculation module is used to calculate the weight of each picture channel, and the voting judgment module is used to vote and judge whether the target occurs;

[0041] The feature template management module includes: a history record management unit, a feature extraction unit and a feature template storage unit, wherein the history record management unit is used to manage the history records of the multi-source device display screen, the feature extraction unit is used to extract the feature screen of the target event, and the feature template storage unit is used to store the feature template of the target event;

[0042] The discrete value calculation module includes: a picture acquisition unit and a feature comparison unit, wherein the picture acquisition unit is used to acquire pictures in the picture display channel, and the feature comparison unit is used to compare the acquired sampling data with the feature template;

[0043] The circular sampling module includes: a discrete calculation unit, a discreteness judgment unit and a supplementary sampling unit, wherein the discrete calculation unit is used to calculate the discrete value of the similarity, the discreteness judgment unit is used to compare whether the discreteness meets the output condition, and the supplementary sampling unit is used to perform supplementary sampling on the display picture;

[0044] The weight calculation module includes: a sampling statistics unit and a weight calculation unit, wherein the sampling statistics unit is used to count the total number of samples of the screen display channel in a unit sampling period, and the weight calculation unit is used to calculate the weight value of the screen display channel;

[0045] The voting judgment module includes: an average value calculation unit, a voting coefficient calculation unit and a comparison judgment unit, wherein the average value calculation unit is used to calculate the average value of the similarity, the voting coefficient calculation unit is used to calculate the voting coefficient corresponding to the screen display channel, and the comparison judgment unit is used to judge whether the target event occurs.

[0046] Compared with the prior art, the beneficial effect achieved by the present invention is as follows: the present invention compares the collected images with the feature templates, compares each image display channel with the feature template, performs distributed judgment on whether the target event occurs in each image channel, and then integrates the distributed judgment results, thereby improving the accuracy of the judgment and assisting relevant management personnel in monitoring multi-channel data sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0048] Figure 1 This is a schematic diagram of the structure of a multi-source device display process supervision system based on the Internet of Things in the present invention;

[0049] Figure 2 This is a flow chart of a multi-source device display process supervision method based on the Internet of Things in the patent of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] See also Figure 1 and Figure 2 , the present invention provides a technical solution:

[0052] Step S100: collecting display images of various data sources of a multi-source display device, setting the display images as target events when they meet a certain feature, collecting historical records of the multi-source display device in displaying the target event, and compiling them into a feature template;

[0053] Wherein, step S100 includes:

[0054] Step S101: at least two data sources are connected to the multi-source display device, wherein each data source corresponds to a picture display channel, and information collected from each data source is displayed in the corresponding picture display channel;

[0055] Step S102: acquiring the historical records displayed by each screen display channel when the target event occurs, and extracting features from the historical records displayed by each screen display channel to obtain feature screens corresponding to each screen display channel when the target event occurs;

[0056] Step S103: all picture display channels in the multi-source display device are arranged in order, and according to the arrangement order of the picture display channels, characteristic pictures corresponding to the picture display channels are collected to obtain characteristic templates corresponding to the target event;

[0057] Step S200: sampling the real-time display pictures of each data source of the multi-source display device, calculating the similarity between the sampled display pictures and the feature template, sampling the real-time display pictures of each data source several times, and calculating the discrete values ​​of the similarity of the same data source respectively;

[0058] Wherein, step S200 includes:

[0059] Step S201: setting a unit sampling period with a time length of T0 and a unit sampling number a, sampling the pictures displayed in each picture display channel of the multi-source display device after entering any unit sampling period to obtain a sampled picture corresponding to each picture display channel, wherein the number of sampled pictures of each picture display channel is a;

[0060] Step S202: comparing the sampled images of each image display channel with the characteristic images of the corresponding image display channel to obtain the similarity between each sampled image and the characteristic image, and classifying and collecting the similarities according to the corresponding image display channel to obtain a first similarity data group corresponding to each image display channel;

[0061] Step S203: calculating the discrete values ​​corresponding to the data in the first similarity data group of each picture display channel, and matching the discrete values ​​with the picture display channels;

[0062] In the embodiment, graphic similarity and data similarity are used, wherein, when the content displayed in the picture display channel is an image, such as a monitoring image or a signal waveform graph, the similarity between the sampled picture and the feature picture is calculated by comparing the graphic similarity of the image;

[0063] When the similarity between the sampling screen and the feature screen is data, the boundary conditions for triggering the target event are obtained. For example, when the value exceeds bor, the target event occurs, and the value value of the screen display channel is obtained. When value ≥ bor, the similarity β = 1, and when value < bor, the similarity β = (bor-value) / bor;

[0064] Step S300: setting a discreteness evaluation threshold, when the discreteness value of the real-time display screen similarity of the data source is greater than the discreteness evaluation threshold, continuing sampling, when the similarity between the newly added sampled display screen and the feature template makes the discreteness value of the similarity decrease, stopping further sampling;

[0065] Wherein, step S300 includes:

[0066] Step S301: Compare the discrete values ​​corresponding to each screen display channel with the discreteness evaluation threshold, obtain the screen display channels corresponding to the discrete values ​​greater than or equal to the discreteness evaluation threshold, and merge the screen display channels into a to-be-detected group, wherein the discrete value corresponding to the i-th screen display channel in the to-be-detected group is recorded as qi;

[0067] Step S302: for the picture display channel in the to-be-detected group, a number of sampled pictures are collected again, and similarities between the sampled pictures and corresponding feature pictures are calculated respectively, to obtain a second similarity data group corresponding to the picture display channel in the to-be-detected group;

[0068] Step S303: obtaining all data in the first similarity data group and the second similarity group of the i-th screen display channel in the group to be detected, calculating the discrete value of all the data and recording it as q'i, and when q'i≤qi, proceeding to step S400;

[0069] Step S304: when q'i>qi, the i-th picture display channel in the to-be-detected group is re-integrated into the to-be-detected group, and q'i is used as the new qi to repeat steps S301 to S302 until the output condition of step S303 is met and the loop is stopped;

[0070] Step S400: Calculate the weight value of the real-time display screen of each data source according to the sampling times corresponding to each data source, wherein the weight value is inversely proportional to the sampling times;

[0071] Wherein, step S400 includes:

[0072] Step S401: Obtain the total number of sampling times corresponding to each picture display channel of the multi-source display device, wherein the total number of sampling times corresponding to the j-th picture display channel in the multi-source display device is recorded as NA j , the weight coefficient corresponding to the j-th screen display channel is recorded as μ j , μ j =a / NA j ;

[0073] Step S402: setting the initial weight value k0, the weight value corresponding to the j-th picture display channel in the multi-source display device is p j , pj =k0×μ j ;

[0074] Step S500: Calculate the weight of the real-time display screen of each data source when judging whether the target event occurs according to the weight value, use the weight as the voting coefficient corresponding to each data source, and each display screen votes on whether the target event occurs to judge whether the target event occurs;

[0075] Wherein, step S500 includes:

[0076] Step S501: obtaining r historical records of comparison between a sampled image and a corresponding feature image of any image display channel of the multi-source display device, wherein one historical record corresponds to one comparison between the sampled image and the corresponding feature image, and merging the similarities in the r historical records into a set R;

[0077] Step S502: Obtain the maximum similarity value e in R max and the minimum similarity e min , set the similarity reference range (c, d), where c < e min , d>e max ;

[0078] Preferably, when the similarity is in the interval range of (0,1), c is 0 and d is 1;

[0079] Step S503: Calculate the average similarity values ​​corresponding to the image display channels of the multi-source display device, wherein the average similarity value corresponding to the j-th image display channel in the multi-source display device is ave j , where h m represents the mth similarity of the jth picture display channel;

[0080] Step S504: Calculate the voting coefficient vote corresponding to the j-th picture display channel j ,

[0081] Step S505: Calculate the total number of votes V of all picture display channels in the multi-source display device. Among them, vote f represents the voting coefficient corresponding to the f-th picture display channel of the multi-source display device, and g represents the total number of picture display channels included in the multi-source display device;

[0082] In the embodiment, there are 4 picture display channels, namely CH1, CH2, CH3 and CH4. When a=3, each picture channel is sampled, and the sampling results are:

[0083] CH1: 0.7, 0.72, 0.69; CH2: 0.82, 0.45, 0.6, 0.9, 0.3, 0.7;

[0084] CH3: 0.3, 0.28, 0.32; CH4: 0.42, 0.58, 0.5;

[0085] Set the initial weight value k0=1, the corresponding weight values ​​of CH1, CH3 and CH4 are 1, and the weight value of CH2 is 0.5;

[0086] The voting coefficient of CH1 is vote1 = ((0.7 + 0.72 + 0.69) / 3) - 0.5 = 0.203;

[0087] The voting coefficient of CH2 is vote2=0.5×(((0.82+0.45+0.6+0.9+0.3+0.7) / 6)-0.5)=0.06;

[0088] The voting coefficient of CH3 is vote3 = ((0.3 + 0.28 + 0.32) / 3) - 0.5 = -0.2;

[0089] The voting coefficient of CH4 is vote4=((0.42+0.58+0.5) / 3)-0.5=0;

[0090] V=0.23+0.06-0.2+0=0.09; it is judged that the target event occurs;

[0091] Step S506: When V>0, it is determined that the target event has occurred, and when V≤0, it is determined that the target event has not occurred, and information reminders are given to relevant personnel according to the determination results.

[0092] The system includes:

[0093] Feature template management module, discrete value calculation module, cyclic sampling module, weight calculation module and voting judgment module;

[0094] The feature template management module is used to manage the feature template of the target event, wherein the feature template management module includes: a history record management unit, a feature extraction unit and a feature template storage unit, wherein the history record management unit is used to manage the history records of the multi-source device display screen, the feature extraction unit is used to extract the feature screen of the target event, and the feature template storage unit is used to store the feature template of the target event;

[0095] The discrete value calculation module is used to calculate the discrete value of the similarity, the picture acquisition unit is used to acquire the pictures in the picture display channel, and the feature comparison unit is used to compare the acquired sample data with the feature template;

[0096] The cyclic sampling module is used to perform cyclic control on the picture sampling, wherein the cyclic sampling module includes: a discrete calculation unit, a discreteness judgment unit and a supplementary sampling unit, wherein the discrete calculation unit is used to calculate the discrete value of the similarity, the discreteness judgment unit is used to compare whether the discreteness meets the output condition, and the supplementary sampling unit is used to perform supplementary sampling on the display picture;

[0097] The weight calculation module is used to calculate the weight of each picture channel, wherein the weight calculation module includes: a sampling statistics unit and a weight calculation unit, wherein the sampling statistics unit is used to count the total number of samples of the picture display channel in a unit sampling cycle, and the weight calculation unit is used to calculate the weight value of the picture display channel;

[0098] Among them, the voting and judgment module is used to vote and judge whether the target occurs, wherein the voting and judgment module includes: an average value calculation unit, a voting coefficient calculation unit and a comparison and judgment unit, wherein the average value calculation unit is used to calculate the average value of the similarity, the voting coefficient calculation unit is used to calculate the voting coefficient corresponding to the screen display channel, and the comparison and judgment unit is used to judge whether the target event occurs.

[0099] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0100] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A multi-source device display process supervision method based on the Internet of Things, It is characterized in that The method comprises the following steps: Step S100: collecting display images of various data sources of a multi-source display device, setting the display images as target events when they meet a certain feature, collecting historical records of the multi-source display device in displaying the target event, and compiling them into a feature template; Step S200: sampling the real-time display pictures of each data source of the multi-source display device, calculating the similarity between the sampled display pictures and the feature template, sampling the real-time display pictures of each data source several times, and calculating the discrete values ​​of the similarity of the same data source respectively; Step S300: setting a discreteness evaluation threshold, when the discreteness value of the real-time display screen similarity of the data source is greater than the discreteness evaluation threshold, continuing sampling, when the similarity between the newly added sampled display screen and the feature template makes the discreteness value of the similarity decrease, stopping further sampling; Step S400: Calculate the weight value of the real-time display screen of each data source according to the sampling times corresponding to each data source, wherein the weight value is inversely proportional to the sampling times; Step S500: Calculate the weight of the real-time display screen of each data source when judging whether the target event occurs according to the weight value, use the weight as the voting coefficient corresponding to each data source, and each display screen votes on whether the target event occurs to judge whether the target event occurs.

2. According to claim 1, a multi-source device display process supervision method based on the Internet of Things, Features: Step S100 includes: Step S101: at least two data sources are connected to the multi-source display device, wherein each data source corresponds to a picture display channel, and information collected from each data source is displayed in the corresponding picture display channel; Step S102: acquiring the historical records displayed by each screen display channel when the target event occurs, and extracting features from the historical records displayed by each screen display channel to obtain feature screens corresponding to each screen display channel when the target event occurs; Step S103: all picture display channels in the multi-source display device are arranged in order, and according to the arrangement order of the picture display channels, characteristic pictures corresponding to the respective picture display channels are collected to obtain characteristic templates corresponding to the target event.

3. According to claim 2, a multi-source device display process supervision method based on the Internet of Things, Features: Step S200 includes: Step S201: Set the time length to T 0 A unit sampling period and a unit sampling number a are provided, and after entering any unit sampling period, the pictures displayed in each picture display channel of the multi-source display device are sampled to obtain sampled pictures corresponding to each picture display channel, wherein the number of sampled pictures of each picture display channel is a; Step S202: comparing the sampled images of each image display channel with the characteristic images of the corresponding image display channel to obtain the similarity between each sampled image and the characteristic image, and classifying and collecting the similarities according to the corresponding image display channel to obtain a first similarity data group corresponding to each image display channel; Step S203: calculating discrete values ​​corresponding to the data in the first similarity data group of each picture display channel, and matching the discrete values ​​with the picture display channels.

4. According to claim 3, a multi-source device display process supervision method based on the Internet of Things, Features: Step S300 includes: Step S301: Compare the discrete values ​​corresponding to each screen display channel with the discreteness evaluation threshold, obtain the screen display channels corresponding to the discrete values ​​greater than or equal to the discreteness evaluation threshold, and merge the screen display channels into a to-be-detected group, wherein the discrete value corresponding to the i-th screen display channel in the to-be-detected group is recorded as qi; Step S302: for the picture display channel in the to-be-detected group, a number of sampled pictures are collected again, and similarities between the sampled pictures and corresponding feature pictures are calculated respectively, to obtain a second similarity data group corresponding to the picture display channel in the to-be-detected group; Step S303: obtaining all data in the first similarity data group and the second similarity group of the i-th screen display channel in the group to be detected, calculating the discrete value of all the data and recording it as q'i, and when q'i≤qi, proceeding to step S400; Step S304: When q'i>qi, the i-th picture display channel in the to-be-detected group is reintegrated into the to-be-detected group, and q'i is used as the new qi to repeat steps S301 to S302 until the output condition of step S303 is met and the loop is stopped.

5. According to claim 4, a multi-source device display process supervision method based on the Internet of Things, Features: Step S400 includes: Step S401: Obtain the total number of sampling times corresponding to each picture display channel of the multi-source display device, wherein the total number of sampling times corresponding to the j-th picture display channel in the multi-source display device is recorded as NA j , the weight coefficient corresponding to the j-th screen display channel is recorded as μ j , μ j =a / NA j ; Step S402: Setting the initial weight value k 0 , the weight value corresponding to the j-th picture display channel in the multi-source display device is p j , p j =k 0 ×μ j .

6. According to claim 5, a multi-source device display process supervision method based on the Internet of Things, Features: Step S500 includes: Step S501: obtaining r historical records of comparison between a sampled image and a corresponding feature image of any image display channel of the multi-source display device, wherein one historical record corresponds to one comparison between the sampled image and the corresponding feature image, and merging the similarities in the r historical records into a set R; Step S502: Obtain the maximum similarity value e in R max and the minimum similarity e min , set the similarity reference range (c, d), where c < e min , d>e max ; Step S503: Calculate the average similarity values ​​corresponding to the image display channels of the multi-source display device, wherein the average similarity value corresponding to the j-th image display channel in the multi-source display device is ave j , where h m represents the mth similarity of the jth picture display channel; Step S504: Calculate the voting coefficient vote corresponding to the j-th picture display channel j , Step S505: Calculate the total number of votes V of all picture display channels in the multi-source display device. Among them, vote f represents the voting coefficient corresponding to the f-th picture display channel of the multi-source display device, and g represents the total number of picture display channels included in the multi-source display device; Step S506: When V>0, it is determined that the target event has occurred, and when V≤0, it is determined that the target event has not occurred, and information reminders are given to relevant personnel according to the determination results.

7. A multi-source device display process supervision system applied to the multi-source device display process supervision method based on the Internet of Things as described in any one of claims 1 to 6, It is characterized in that The system includes the following modules : Feature template management module, discrete value calculation module, cyclic sampling module, weight calculation module and voting judgment module, wherein the feature template management module is used to manage the feature template of the target event, the discrete value calculation module is used to calculate the discrete value of the similarity, the cyclic sampling module is used to cyclically control the picture sampling, the weight calculation module is used to calculate the weight of each picture channel, and the voting judgment module is used to vote and judge whether the target occurs.

8. The multi-source device display process monitoring system according to claim 7, Features: The feature template management module includes: a history record management unit, a feature extraction unit and a feature template storage unit, wherein the history record management unit is used to manage the history records of the multi-source device display screen, the feature extraction unit is used to extract the feature screen of the target event, and the feature template storage unit is used to store the feature template of the target event; The discrete value calculation module includes: a picture acquisition unit and a feature comparison unit, wherein the picture acquisition unit is used to acquire pictures in the picture display channel, and the feature comparison unit is used to compare the acquired sampling data with the feature template.

9. The multi-source device display process monitoring system according to claim 8, Features: The cyclic sampling module includes: a discrete calculation unit, a discreteness judgment unit and a supplementary sampling unit, wherein the discrete calculation unit is used to calculate the discrete value of the similarity, the discreteness judgment unit is used to compare whether the discreteness meets the output condition, and the supplementary sampling unit is used to perform supplementary sampling on the display picture.

10. The multi-source device display process monitoring system according to claim 9, Features: The weight calculation module includes: a sampling statistics unit and a weight calculation unit, wherein the sampling statistics unit is used to count the total number of samples of the screen display channel in a unit sampling period, and the weight calculation unit is used to calculate the weight value of the screen display channel; The voting judgment module includes: an average value calculation unit, a voting coefficient calculation unit and a comparison judgment unit, wherein the average value calculation unit is used to calculate the average value of the similarity, the voting coefficient calculation unit is used to calculate the voting coefficient corresponding to the screen display channel, and the comparison judgment unit is used to judge whether the target event occurs.

Images