A method, system and electronic device for monitoring human-machine interface factor saliency adjustment
Patent Information
- Application Number
- CN202310952250.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-31
- Publication Date
- 2026-09-04
- Estimated Expiration
- 2043-07-31
AI Technical Summary
然而在信息增强的人机协作过程中，很难实现界面元素显著性与任务价值的最佳同步，即最显著的元素可能不是当前任务中最需要观察的数据
[0035]To address the problem of low monitoring efficiency caused by excessive information in complex monitoring human-machine interfaces, this invention establishes a novel monitoring human-machine interface that integrates top-down and bottom-up attention channels. This novel monitoring human-machine interface has the characteristic of automatically adjusting the salience of interface elements, resulting in better visual search performance in actual monitoring tasks and reducing the risk of system failure.
Smart Images
Figure CN117130703B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of monitoring systems and human-machine interface technology, and in particular to a method, system and electronic device for adjusting the salience of monitoring human-machine interface factors based on bottom-up attention channels and top-down attention channels. Background Technology
[0002] With the rapid development of computer information technology, the human-machine interface (HMI) of monitoring systems (hereinafter referred to as the monitoring HMI) is showing a trend of information integration. On the one hand, the integrated display of information greatly improves the automation level of the monitoring HMI and enhances its data processing capabilities. On the other hand, the integrated display of information leads to a sharp increase in the amount of monitoring information, increasing the complexity of multi-task, multi-information HMI monitoring tasks. The excessive information on the monitoring HMI, leading to problems such as difficulty in capturing information or information overload, poses significant security risks to the monitoring system.
[0003] Existing human-machine interfaces for monitoring generally lack dynamic enhancement technology, and the salience of interface elements is set at the initial design stage. However, in the process of information-enhanced human-machine collaboration, it is difficult to achieve optimal synchronization between the salience of interface elements and the value of the task; that is, the most salient element may not be the data that is most needed to be observed in the current task. Summary of the Invention
[0004] The purpose of this invention is to provide a method, system, and electronic device for adjusting the salience of monitoring human-machine interface factors, which can automatically adjust the salience of monitoring human-machine interface elements.
[0005] To achieve the above objectives, the present invention provides the following solution:
[0006] In a first aspect, the present invention provides a method for adjusting the salience of monitoring human-machine interface factors, comprising:
[0007] The monitoring human-machine interface is divided into multiple functional modules;
[0008] Calculate the inherent information value of each of the aforementioned functional modules;
[0009] The inherent information value of each functional module is combined with the image saliency of the monitoring human-machine interface to determine the average saliency of each functional module under the bottom-up attention channel.
[0010] Calculate the feature information related to the monitoring task for each of the functional modules, and determine the average salience of each of the functional modules in the top-down attention channel based on the inherent information value and the feature information.
[0011] Based on the average saliency of each functional module under the bottom-up attention channel and the average saliency of each functional module under the top-down attention channel, the judgment coefficient of each functional module is calculated, and the saliency of monitoring human-machine interface factors is adjusted according to the judgment coefficient of each functional module.
[0012] Optionally, the inherent information value of each functional module is combined with the image saliency of the monitoring human-machine interface to determine the average saliency of each functional module under the bottom-up attention channel, specifically including:
[0013] Calculate the image saliency of the monitoring human-machine interface; the image saliency has 4 channels, namely pixel intensity channel, pixel color channel, pixel orientation channel and other channels;
[0014] Based on the inherent information value of the functional module, the other channels at the corresponding locations are updated to obtain the updated other channels, and the image salience of the monitoring human-machine interface is updated based on the updated other channels.
[0015] Based on the image saliency of the updated monitoring human-machine interface, the average saliency of each functional module is calculated in the bottom-up attention channel.
[0016] Optionally, the feature information includes information entropy and security value.
[0017] Optionally, based on the inherent information value and the feature information, the average salience of each functional module under the top-down attention channel is determined, specifically including:
[0018] The entropy weight method is used to determine the weights corresponding to the inherent information value, the information entropy, and the security value.
[0019] Based on the inherent information value and its corresponding weight, the information entropy and its corresponding weight, and the security value and its corresponding weight, a weighted summation method is used to determine the average saliency of each functional module under the top-down attention channel.
[0020] Optionally, the judgment coefficient of each functional module is calculated based on the average saliency of each functional module in the bottom-up attention channel and the average saliency of each functional module in the top-down attention channel, specifically including:
[0021] Using mathematical statistics theory, linear regression is performed on the average significance of each functional module under the bottom-up attention channel and the average significance of each functional module under the top-down attention channel to determine the judgment coefficient of each functional module.
[0022] Optionally, the salience of monitoring human-machine interface factors is adjusted based on the judgment coefficient of each functional module, specifically including:
[0023] Determine whether the judgment coefficient of the functional module is greater than or equal to a set value;
[0024] If not, the prominence of the functional module will be adjusted.
[0025] Optionally, the prominence of the functional modules can be adjusted, specifically including:
[0026] The salience of the functional module is adjusted by using a pixel color intensity adjustment method.
[0027] Secondly, the present invention provides a system for monitoring the salience adjustment of human-machine interface factors, comprising:
[0028] The module division is used to divide the monitoring human-machine interface into multiple functional modules;
[0029] An inherent information value calculation module is used to calculate the inherent information value of each of the aforementioned functional modules;
[0030] The bottom-up average saliency calculation module is used to combine the inherent information value of each functional module with the image saliency of the monitoring human-machine interface to determine the average saliency of each functional module under the bottom-up attention channel.
[0031] The top-down average saliency calculation module is used to calculate the feature information of each functional module related to the monitoring task, and determine the average saliency of each functional module under the top-down attention channel based on the inherent information value and the feature information.
[0032] An adjustment module is used to calculate the judgment coefficient of each functional module based on the average salience of each functional module in the bottom-up attention channel and the average salience of each functional module in the top-down attention channel, and to adjust the salience of monitoring human-machine interface factors based on the judgment coefficient of each functional module.
[0033] Thirdly, the present invention provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to cause the electronic device to perform the monitoring human-machine interface factor saliency adjustment method according to the first aspect.
[0034] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:
[0035] To address the problem of low monitoring efficiency caused by excessive information in complex monitoring human-machine interfaces, this invention establishes a novel monitoring human-machine interface that integrates top-down and bottom-up attention channels. This novel monitoring human-machine interface has the characteristic of automatically adjusting the salience of interface elements, resulting in better visual search performance in actual monitoring tasks and reducing the risk of system failure. Attached Figure Description
[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0037] Figure 1 A flowchart illustrating the method for adjusting the saliency of human-machine interface factors provided in an embodiment of the present invention;
[0038] Figure 2 This is a schematic diagram of the HMI (Human-Machine Interface) division of a subway vehicle-mounted ATP (Automatic Train Protection) system provided in an embodiment of the present invention;
[0039] Figure 3 This is a flowchart illustrating the calculation of the average saliency of functional modules under the bottom-up attention channel, as provided in an embodiment of the present invention.
[0040] Figure 4 A schematic diagram of the bottom-up saliency calculation results of the HMI human-machine interface of a subway vehicle ATP provided in an embodiment of the present invention;
[0041] Figure 5 This is a flowchart illustrating the calculation of the average saliency of functional modules under the top-down attention channel, as provided in an embodiment of the present invention.
[0042] Figure 6 A flowchart for adjusting the saliency of interface factors provided in an embodiment of the present invention;
[0043] Figure 7 This is a schematic diagram of the structure of the human-machine interface factor saliency adjustment system provided in an embodiment of the present invention. Detailed Implementation
[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0045] According to the visual processing theory of attention, human attention processing tools have two channels: a rapid, bottom-up saliency-driven approach and a slower, top-down task-dependent approach. Bottom-up saliency attention to information is a rapid and short-lived process, driven by visual features. Top-down task-driven behavior is a long-term and stable process, driven by task features such as urgency, expectation, effort, and value in monitoring tasks. However, in complex monitoring human-computer interfaces, the driving results of these two channels are often inconsistent. For example, an element with high interface saliency (a prominent red element) may have high bottom-up saliency, but sometimes its task urgency is not high, meaning its top-down saliency is low. This mismatch between the two channels makes it difficult for users of monitoring human-computer interfaces to obtain data with maximum task value in a timely manner, resulting in impaired task performance. Therefore, this invention dynamically adjusts the saliency of bottom-up interface factors during the operation of monitoring tasks to improve the matching degree between bottom-up and top-down saliency, thereby enhancing the perceptual processing efficiency of monitoring human-computer interface users.
[0046] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0047] Example 1
[0048] like Figure 1 As shown, this embodiment provides a method for adjusting the saliency of monitoring human-computer interface factors based on bottom-up attention channels and top-down attention channels, including:
[0049] Step 101: Divide the monitoring human-machine interface into multiple functional modules.
[0050] The division of the above functional modules should be based on full consultation with experts or reference to the interface design documents. Figure 2 This is an example of the HMI (Human-machine Interface) partitioning for a subway onboard ATP (Automatic Train Protection) system, where A1, A2, B, etc., are the partitioned functional modules.
[0051] The principles for dividing functional modules are: (1) Functional modules should have independent functions, such as train speed display modules. (2) All functional modules should include the entire interface area, that is, the collection of all functional modules should be a complete monitoring human-machine interface.
[0052] Step 102: Calculate the inherent information value of each of the functional modules.
[0053] In this embodiment, step 102 specifically includes:
[0054] The inherent information value V (ranging from 0 to 1, where 0 represents unimportant and 1 represents very important) of each functional module is determined through expert questionnaire consultation. This inherent information value V represents the static importance of the functional module in the monitoring system, which is a top-down module characteristic related to the monitoring task.
[0055] Specifically, the algorithm for data consistency testing is used to process the information value questionnaires given by multiple experts to obtain the inherent information value of each functional module. This testing method generally uses Cronbach's alpha, which is required to be greater than or equal to 0.7.
[0056] Step 103: Combine the inherent information value of each functional module with the image saliency of the monitoring human-machine interface to determine the average saliency of each functional module under the bottom-up attention channel.
[0057] In this embodiment, step 103 specifically includes:
[0058] First, the image saliency of the monitoring human-machine interface is calculated. The image saliency has four channels: pixel intensity channel, pixel color channel, pixel orientation channel, and other channels. Second, based on the inherent information value of the functional modules, the other channels at the corresponding positions are updated to obtain the updated other channels. Based on the updated other channels, the image saliency of the monitoring human-machine interface is updated. Finally, based on the updated image saliency of the monitoring human-machine interface, the average saliency of each functional module under the bottom-up attention channel is calculated.
[0059] Furthermore, the specific operation of step 103 is as follows:
[0060] a. Calculate the image saliency of the monitoring human-machine interface according to the Itti image saliency algorithm (IttiL / Koch2001); the image saliency has 4 channels, namely pixel intensity I (Intensity) channel, pixel color C (Colours) channel, pixel orientation O (Orientations) channel, and other T (Other) channel.
[0061] Wherein, image saliency S (Salency) can be quantified as
[0062] In the original model for calculating the saliency of Itti images, the [Other T] channels typically incorporate motion (M) or flicker (F). Considering the characteristics of monitoring human-machine interfaces, the inherent information value (V) of functional modules is also incorporated here, that is, the top-down module features are integrated to correct the bottom-up saliency values.
[0063] The updated [Other T] channels are as follows:
[0064] Image saliency S is a pixel-based numerical value. For example, in an interface with a resolution of 1920*1080, the value of S is Si. x-y The dataset contains 1920 x-1920 and 1080 y-1080 data points. Each functional module contains a number of pixels. The average saliency of each functional module is calculated. This value represents the average significance of the final bottom-up functional modules.
[0065] The formula for calculating the average significance of the i-th functional module is as follows:
[0066]
[0067] The Avr() function is an average value function.
[0068] Figure 3 This describes the calculation process for the average saliency of functional modules under the bottom-up attention channel. Figure 4 This is the bottom-up saliency calculation result of the HMI (Human-Machine Interface) of a subway car's onboard ATP (Automatic Train Protection) system. Figure 4 In the middle, the brighter parts have more prominent module features from bottom to top.
[0069] Step 104: Calculate the feature information related to the monitoring task for each functional module, and determine the average salience of each functional module under the top-down attention channel based on the inherent information value and the feature information. The feature information includes information entropy and security value.
[0070] The specific process of step 104 is as follows: Figure 5 As shown.
[0071] In this embodiment, the importance of a functional module is defined by its inherent information value V. A functional module with a higher inherent information value V has higher top-down salience, and users should allocate their attention to functional modules with high inherent information value V.
[0072] In this embodiment, the information entropy H of the functional module is the expected value of the amount of information carried by the functional module (ranging from 0 to 1), and its calculation formula is H = E[-logp] i Where E[] is the mathematical expectation, P i Let be the information probability of the i-th functional module.
[0073] For example, the emergency braking function module has only two states: not braking and braking. The information probabilities of these two states are 40% and 60%, respectively. That is, the information probability of this function module is P0 = 40% or P1 = 60%.
[0074] The information probability of each functional module can be derived from long-term system data or determined by the evaluators through expert interviews. Information entropy is used to characterize the uncertainty of a functional module. The higher the uncertainty, the more detrimental it is to the monitoring system. Users should focus their attention on functional modules with higher information entropy in order to eliminate the uncertainty of the monitoring human-machine interface as quickly as possible.
[0075] In this embodiment, the safety value D of the functional module is: the information in the functional module generally has a safe range, for example, the train speed must not exceed 80km / h, etc. If the information value in the current functional module is D... n (Current vehicle speed 70km / h), which is close to the safety limit value D. l Between (80km / h), there is a safety margin |D n -D l |(10km / h), similarly, the maximum value of the remaining safety margin D max (Maximum possible range: 0km to 80km, i.e., 80km / h) can be derived from long-term system data or determined by the evaluators through expert interviews. This can then be further determined using a formula. After normalization, the safety margin D' of the functional module can be obtained, and the value range of the safety margin D' of the functional module is also 0-1.
[0076] The safety margin D' is a negative indicator, meaning that the smaller the safety margin of a functional module, the more attention users should allocate to that module. Therefore, the safety margin D' is modified to a positive indicator, namely the safety value D, which is the complement of the safety margin, and the calculation formula is D = 1 - D'. Thus, the higher the safety value (ranging from 0 to 1), the more attention users should allocate to that functional module.
[0077] The importance (V), information entropy (H), and security value (D) of functional modules—these three top-down features of functional modules—all have a positive impact on the top-down attention saliency; that is, the larger their values, the higher the average saliency of the corresponding top-down functional modules. The larger the value, the better. The weights of the three features can be obtained using the traditional entropy weight method, denoted as w. V w H w D After weighted summation, the result can be obtained.
[0078] The formula for calculating the average saliency of the i-th functional module under the top-down attention channel is as follows:
[0079] Step 105: Calculate the judgment coefficient of each functional module based on the average salience of each functional module in the bottom-up attention channel and the average salience of each functional module in the top-down attention channel, and adjust the salience of monitoring human-machine interface factors based on the judgment coefficient of each functional module.
[0080] The specific process of step 105 is as follows: Figure 6 As shown.
[0081] In this embodiment, the judgment coefficient of each functional module is calculated based on the average salience of each functional module under the bottom-up attention channel and the average salience of each functional module under the top-down attention channel, specifically including:
[0082] Using mathematical statistics theory, linear regression was performed on the average significance of each functional module under the bottom-up attention channel and the average significance of each functional module under the top-down attention channel to determine the judgment coefficient r of each functional module. 2 .
[0083] In this embodiment, the salience of monitoring human-machine interface factors is adjusted according to the judgment coefficient of each functional module, specifically including:
[0084] For any functional module, determine whether the judgment coefficient of the functional module is greater than or equal to a set value; if not, adjust the significance of the functional module; if yes, do not adjust the significance of the functional module.
[0085] For example, when r 2 When the saliency is less than 0.8, it is considered that the linear correlation between the first saliency (i.e., the average saliency of the functional modules under the bottom-up attention channel) and the second saliency (i.e., the average saliency of the functional modules under the top-down attention channel) is poor. In other words, the elements that are prominent on the interface (such as red elements) do not have the value of the corresponding prominent work task. Observing this element is a waste of the user's attention to some extent. At this time, the saliency of the corresponding elements in the human-computer interface should be adjusted.
[0086] In the above linear regression calculation process, discrete points can be obtained, which represent the functional modules that deviate significantly from linearity. That is, these functional modules have the worst significance matching from top to bottom and bottom to top.
[0087] Considering that other bottom-up parameters of the human-computer interface elements are difficult to adjust quickly, the pixel color intensity of this functional module was adjusted to eliminate the mismatch. Specifically, a color adjustment amount Δ was added to the pixel color intensity. C Color adjustment amount Δ C The absolute value decreases sequentially according to the degree of deviation. In a monitoring human-machine interface with n functional modules, the module with the largest deviation has a value of n for j, the module with the second largest deviation has a value of n-1 for j, and so on. Thus, the greater the degree of deviation, the greater the color adjustment Δ. C The larger the value, the more color adjustment Δ C The sign depends on whether the linear deviation is positive or negative (i.e., whether the function module should increase color brightness to attract user attention or decrease color brightness to avoid attracting user attention). The obtained color adjustment amount Δ... C Then, the original pixel color intensity C can be corrected to C+Δ. C This allows for the adjustment of the prominence of interface elements, i.e., functional modules.
[0088] Among them, the color adjustment amount Δ C The calculation formula is:
[0089] To avoid making the interface overly cluttered by adjusting too many interface elements at the same time, the color intensity of only one discrete point is adjusted at a time, and then the bottom-up average significance is recalculated. Recalculate r for linear regression 2 Value. If r 2 If the value is less than 0.8, the process is repeated to adjust the next discrete point until r is verified. 2 When the value is >=0.8, it is believed that the salience of the bottom-up and top-down interfaces has a good linear correlation, that is, the two cognitive channels have a unified attention direction, and users can easily perceive the most important functional module in the current task in the human-computer interface, and then complete the interface adjustment.
[0090] Through the steps described above, a novel monitoring human-machine interface that integrates top-down and bottom-up attention channels has been constructed. This interface can match the two attention channels, ensuring that the most prominent elements on the interface are also the most important elements in the monitoring task. This human-machine interface can greatly enhance the user's situational awareness and operational performance, and can also improve the system's security to a certain extent.
[0091] Example 2
[0092] In order to implement the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, a monitoring system for adjusting the salience of human-machine interface factors is provided below.
[0093] like Figure 7 The present embodiment provides a system for monitoring the salience adjustment of human-machine interface factors, including:
[0094] The partitioning module 701 is used to partition the monitoring human-machine interface into multiple functional modules.
[0095] The inherent information value calculation module 702 is used to calculate the inherent information value of each of the functional modules.
[0096] Bottom-up average saliency calculation module 703 is used to combine the inherent information value of each functional module with the image saliency of the monitoring human-machine interface to determine the average saliency of each functional module under the bottom-up attention channel.
[0097] The top-down average saliency calculation module 704 is used to calculate the feature information of each functional module related to the monitoring task, and determine the average saliency of each functional module under the top-down attention channel based on the inherent information value and the feature information.
[0098] The adjustment module 705 is used to calculate the judgment coefficient of each functional module based on the average salience of each functional module in the bottom-up attention channel and the average salience of each functional module in the top-down attention channel, and to adjust the salience of monitoring human-machine interface factors based on the judgment coefficient of each functional module.
[0099] Example 3
[0100] This invention provides an electronic device including a memory and a processor. The memory stores a computer program, and the processor runs the computer program to enable the electronic device to perform the human-machine interface factor saliency adjustment method of Embodiment 1.
[0101] Alternatively, the aforementioned electronic device may be a server.
[0102] In addition, embodiments of the present invention also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the monitoring human-machine interface factor saliency adjustment method of Embodiment 1.
[0103] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.
[0104] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.
Claims
1. A method for adjusting the salience of human-machine interface factors, characterized in that, include: The monitoring human-machine interface is divided into multiple functional modules; Calculate the inherent information value of each of the aforementioned functional modules; The inherent information value of each functional module is combined with the image saliency of the monitoring human-machine interface to determine the average saliency of each functional module under the bottom-up attention channel. Specifically, this includes: calculating the image saliency of the monitoring human-machine interface; the image saliency has four channels: pixel intensity channel, pixel color channel, pixel orientation channel, and other channels; updating the other channels at the corresponding positions according to the inherent information value of the functional module to obtain the updated other channels, and updating the image saliency of the monitoring human-machine interface according to the updated other channels; and calculating the average saliency of each functional module under the bottom-up attention channel based on the updated image saliency of the monitoring human-machine interface. Calculate the feature information related to the monitoring task for each of the functional modules, and determine the average salience of each of the functional modules in the top-down attention channel based on the inherent information value and the feature information. Based on the average saliency of each functional module under the bottom-up attention channel and the average saliency of each functional module under the top-down attention channel, the judgment coefficient of each functional module is calculated, and the saliency of monitoring human-machine interface factors is adjusted according to the judgment coefficient of each functional module.
2. The method for adjusting the salience of monitoring human-machine interface factors according to claim 1, characterized in that, The characteristic information includes information entropy and security value.
3. The method for adjusting the salience of monitoring human-machine interface factors according to claim 2, characterized in that, Based on the inherent information value and the feature information, the average saliency of each functional module under the top-down attention channel is determined, specifically including: The entropy weight method is used to determine the weights corresponding to the inherent information value, the information entropy, and the security value. Based on the inherent information value and its corresponding weight, the information entropy and its corresponding weight, and the security value and its corresponding weight, a weighted summation method is used to determine the average saliency of each functional module under the top-down attention channel.
4. The method for adjusting the salience of human-machine interface factors according to claim 1, characterized in that, Based on the average saliency of each functional module under the bottom-up attention channel and the average saliency of each functional module under the top-down attention channel, the judgment coefficient of each functional module is calculated, specifically including: Using mathematical statistics theory, linear regression is performed on the average significance of each functional module under the bottom-up attention channel and the average significance of each functional module under the top-down attention channel to determine the judgment coefficient of each functional module.
5. The method for adjusting the salience of monitoring human-machine interface factors according to claim 1, characterized in that, The salience of monitoring human-machine interface factors is adjusted based on the judgment coefficient of each functional module, specifically including: Determine whether the judgment coefficient of the functional module is greater than or equal to a set value; If not, the prominence of the functional module will be adjusted.
6. The method for adjusting the salience of monitoring human-machine interface factors according to claim 5, characterized in that, Adjusting the prominence of the aforementioned functional modules specifically includes: The salience of the functional module is adjusted by using a pixel color intensity adjustment method.
7. A system for adjusting the salience of human-machine interface factors, characterized in that, include: The module division is used to divide the monitoring human-machine interface into multiple functional modules; An inherent information value calculation module is used to calculate the inherent information value of each of the aforementioned functional modules; The bottom-up average saliency calculation module combines the inherent information value of each functional module with the image saliency of the monitoring human-machine interface to determine the average saliency of each functional module under the bottom-up attention channel. Specifically, it includes: calculating the image saliency of the monitoring human-machine interface; the image saliency has four channels: pixel intensity channel, pixel color channel, pixel orientation channel, and other channels; updating the other channels at corresponding positions based on the inherent information value of the functional module to obtain updated other channels, and updating the image saliency of the monitoring human-machine interface based on the updated other channels; and calculating the average saliency of each functional module under the bottom-up attention channel based on the updated image saliency of the monitoring human-machine interface. The top-down average saliency calculation module is used to calculate the feature information of each functional module related to the monitoring task, and determine the average saliency of each functional module under the top-down attention channel based on the inherent information value and the feature information. An adjustment module is used to calculate the judgment coefficient of each functional module based on the average salience of each functional module in the bottom-up attention channel and the average salience of each functional module in the top-down attention channel, and to adjust the salience of monitoring human-machine interface factors based on the judgment coefficient of each functional module.
8. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to cause the electronic device to perform the monitoring human-machine interface factor saliency adjustment method according to any one of claims 1 to 6.