Intelligent home travel safety monitoring system

Abstract

The invention discloses a smart home travel safety monitoring system, and relates to the technical field of home-based care for the aged, a road network is established, the number and time of pedestrians on the road network are collected by using a monitoring device on a road, and the movement data of the pedestrians on the road network are analyzed by using big data; obtaining a relationship between pedestrians and time on each road in the road network; the position information and destination information of the old people are obtained, the probability that pedestrians pass through each road on the road network when the old people move on the road network is predicted, and the old people are guided to go to the destination to guide the path of the old people going out and select a proper route in combination with the lighting conditions on the road. The old people are helped to walk safely in the external environment; pedestrians on the road are used for carrying out auxiliary monitoring on the condition of the old people, the pedestrians provide help for the old people when the old people fall down, and harm caused by falling down to the old people is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of home-based elderly care, and in particular to an intelligent home travel safety monitoring system. Background Art

[0002] Elderly people often face safety risks such as falls and sudden illnesses while at home, necessitating effective monitoring and early warning systems to protect their safety and quality of life. Guiding elderly people on safe travel, assisting them in choosing appropriate routes, and helping them navigate safely in external environments are pressing challenges. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent home and travel safety monitoring system to solve the problems raised in the prior art.

[0004] To achieve the above-mentioned objectives, the present invention provides the following technical solution: an intelligent home travel safety monitoring system, comprising a monitoring module, a data storage module, a data analysis module, a path planning module and a guidance module; the monitoring module is used to obtain image data on a road network; the data storage module is used to store historical movement data of pedestrians on the road network; the data analysis module is used to analyze the relationship between pedestrians and time on the road network; the path planning module plans paths for the elderly based on the relationship between pedestrians and time on the road network; and the guidance module is used to guide the elderly from a starting point to a destination.

[0005] Specifically, the path planning module also includes a network construction unit, a path planning unit and a storage unit; the network construction unit is used to generate a weighted graph of the road network, including intersections, roads, and starting and destination information of the elderly in the road network; the path planning unit, based on the weighted graph of the road network, uses the Dijkstra algorithm to plan paths for the elderly; the storage unit is used to store the weighted graph information of the road network.

[0006] Specifically, the path planning unit uses the Dijkstra algorithm to perform path planning, including the following steps:

[0007] S1, store the elderly starting node in the set U, and store the remaining nodes except the elderly starting node in the set V. If the elderly starting node is adjacent to other nodes in V, the weight between the elderly starting node and the adjacent nodes is the weight in the weighted graph; if they are not adjacent, the weight between the elderly starting node and the non-adjacent nodes is positive infinity;

[0008] S2, select a node with the smallest weight between it and the elderly person’s starting node from the set V and add it to the set U, and delete it from the set V;

[0009] S3, taking the node added to set U as the midpoint, modify the weights of each node in set V; if the weight of a node in set V passing through the midpoint to reach the elderly's starting node is smaller than the original weight, then modify the weight of set V. The modified weight is the weight of the elderly's starting node and the midpoint plus the weight of the node in V and the midpoint;

[0010] S4, repeat steps S2 and S3 until all nodes are included in set U.

[0011] The data analysis module uses big data to analyze the movement data of pedestrians on the road network to obtain the relationship between the number of pedestrians on each road in the road network and time; predicts the probability of pedestrians passing each road on the road network when the elderly person moves on the road network based on the starting point information and destination information; and uses big data to analyze the movement data of pedestrians on the road network to obtain the relationship between the number of pedestrians on each road in the road network and time, further comprising the following steps:

[0012] S10, taking the intersections on the road network as nodes and each road as an edge, obtains the time points t0 and t1 from the historical data of the i-th node. end The inflow data of the i-th node between them is anum, where t0 represents the time when the elderly start to move, t end represents the maximum time for the elderly to complete the movement; anum=(hn1+hn2+…+hnk) / k, where hn1, hn2,…, hnk represent the nodes at time points t0 and t end The historical personnel inflow data between, k is the number of historical personnel inflow data;

[0013] S11, obtain the historical movement path data of pedestrians entering the road network from the i-th node, let path j Represents the jth moving path of a pedestrian entering the road network from the i-th node, and obtains the probability P of the j-th moving path appearing j , P j =n j / sum i , where n j Indicates that the pedestrian entering the road network from the i-th node follows the path path j Number of moves, sum i represents the total number of pedestrians entering the road network from the i-th node;

[0014] S12, based on the movement path data and inflow data of the pedestrian at the i-th node, obtain the pedestrian entering from the i-th node on the road network at time points t0 and t end The location information between the time points t0 and t1 is obtained based on the existing movement path data of pedestrians on the road network.end The location information between the two nodes; according to the location information between the pedestrians entering from all nodes at time points t0 and t end The location information between the pedestrians at time points t0 and t end The location information between them is used to obtain the unmanned time of each road on the road network, and the weight of the road is obtained according to the unmanned time of the road.

[0015] Specifically, the data analysis module also includes a spatial conversion unit, a first analysis unit, a second analysis unit and a weight calculation unit; the spatial conversion unit is used to convert the image data of the road into a color space to obtain the brightness information of the road; the first analysis unit is used to analyze the relationship between all existing pedestrian conditions and time on the road network; the second analysis unit is used to generate a virtual model of the pedestrian and obtain the relationship between the virtual pedestrian condition and time on the road network; the weight calculation unit is used to obtain the weight of each road on the road network.

[0016] Specifically, the space conversion unit obtains the image data of each road and converts the image data of each road into a linear RGB color space; for each pixel of the image data of each road, the brightness value is calculated, Y=0.2126R+0.7152G+0.0722B, where R is the red channel value of the image data, G is the green channel value of the image data, B is the blue channel value of the image data, and Y is the brightness value of the image data; let the baseline brightness value of the road be Y max , calculate the relative brightness X of the pixels in the image data, X=Y / Y max , calculate the average brightness AX of the image data based on the relative brightness, and obtain the brightness weight x2 of the road based on the average brightness AX.

[0017] Specifically, the first analysis unit obtains the number of pedestrians on the road network at time points t0 and t1 based on the existing movement path data of pedestrians on the road network. end The relationship between the location information and all existing pedestrian conditions and time on the road network includes the following steps:

[0018] Get the historical movement data of all pedestrians from the historical movement data on the road network, and let the uth historical movement path containing the existing movement path of the ith existing pedestrian on the road network be path u , calculate the u-th historical moving path u The probability P u , P u =n u / sum, where n u Indicates the u-th historical moving path path uThe number of occurrences indicates the number of occurrences of all historical movement paths including the existing movement path of the first pedestrian on the road network;

[0019] According to the pedestrian's moving speed data and moving path data, the number of pedestrians at time points t0 and t end The location information between them is based on the moving path u The probability P u , get the I-th pedestrian appearing on the moving path path u The probability of the road on u Based on the pedestrian's movement speed data, the relationship between the status of the first pedestrian on the road network and time is obtained; the relationship between the position of the first pedestrian on all movement paths and the movement time is obtained. If there are intersecting roads on different movement paths of the first pedestrian, and the time when the first pedestrian arrives at the intersecting road along different movement paths intersects, the probabilities of the different movement paths are added and the minimum value is taken between the sum and one to obtain the relationship between the status of the first pedestrian on the road network and time;

[0020] Get all existing pedestrians at time points t0 and t end According to the situation where there are intersecting roads on the different moving paths of the first pedestrian, the same steps are performed to obtain the relationship between the status of all existing pedestrians on the road network and time.

[0021] Specifically, the second analysis unit obtains the pedestrians entering the road network from the i-th node at time points t0 and t end The relationship between the location information and the status and time of all virtual pedestrians on the road network is specifically comprised of the following steps:

[0022] Get time points t0 and t end The personnel inflow data of the i-th node between time points t0 and t end The personnel inflow data of the i-th node is used to obtain the average time interval avet of the personnel inflow of the i-th node, avet=anum / (t end-t0), according to the time interval avet, generate virtual models of m pedestrians and put them into the i-th node of the road network at time points t0+dt, t0+dt+avet, t0+dt+2×avet, ... t0+dt+m×avet. According to the analysis method of the i-th existing pedestrian on the road network, after the virtual models of m pedestrians are put into the i-th node of the road network, the relationship between the status of the virtual pedestrian at the i-th node on the road network and time is obtained. The relationship between the status of the virtual pedestrians at all nodes on the road network and time is obtained, and the relationship between the status of all virtual pedestrians on the road network and time is obtained. Where m is (t end -t0) is the rounding of the time interval avet, dt is the bias, and is set according to the time when no pedestrians flow into the i-th node at time point t0.

[0023] Specifically, obtaining the weight of a road according to the unmanned time of the road further includes the following steps:

[0024] Get time points t0 and t end The unmanned time period of the i-th road on the road network, let ts and te represent the starting point and end point of an unmanned time period, and calculate the pedestrian weight x3 brought to the road by an unmanned time period. , add up the pedestrian weights brought to the road by all unmanned time periods to obtain the pedestrian weight of the i-th road.

[0025] Specifically, the weight of the weighted graph is calculated using the following formula:

[0026] Z=W1×x1+W2×x2+W3×x3, where Z represents the weight, W1, W2, and W3 represent the connection coefficients, and x1 represents the length of the road; x1 and x2 are fixed values ​​that do not change over time, and x3 is a variable value that changes with time t0.

[0027] Compared with the existing technology, the beneficial effects of the present invention are: guiding the elderly who are traveling, choosing a suitable route, and helping the elderly to walk safely in the external environment; using pedestrians on the road to assist in monitoring the conditions of the elderly, and providing assistance to the elderly when they fall, thereby reducing the harm caused by falls to the elderly. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a structural diagram of the intelligent home and travel safety monitoring system of the present invention. DETAILED DESCRIPTION

[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0030] Example: Figure 1 As shown, the present invention provides an intelligent home travel safety monitoring system, including a monitoring module, a data storage module, a data analysis module, a path planning module and a guidance module; the monitoring module is used to obtain image data on the road network; the data storage module is used to store historical movement data of pedestrians on the road network; the data analysis module is used to analyze the relationship between pedestrians and time on the road network; the path planning module plans paths for the elderly based on the relationship between pedestrians and time on the road network; the guidance module is used to guide the elderly from a starting point to a destination.

[0031] The path planning module also includes a network construction unit, a path planning unit and a storage unit; the network construction unit is used to generate a weighted graph of the road network, including intersections, roads, and starting and destination information of the elderly in the road network; the path planning unit, based on the weighted graph of the road network, uses the Dijkstra algorithm to plan paths for the elderly; and the storage unit is used to store the weighted graph information of the road network.

[0032] The data analysis module uses big data to analyze the movement data of pedestrians on the road network to obtain the relationship between pedestrians and time on each road in the road network; based on the starting point information and destination information of the elderly, it predicts the probability of pedestrians passing each road on the road network when the elderly move on the road network.

[0033] The data analysis module also includes a spatial conversion unit, a first analysis unit, a second analysis unit and a weight calculation unit; the spatial conversion unit is used to convert the image data of the road into a color space to obtain the brightness information of the road; the first analysis unit is used to analyze the relationship between the conditions of all existing pedestrians on the road network and time; the second analysis unit is used to generate a virtual model of the pedestrian and obtain the relationship between the conditions of the virtual pedestrians on the road network and time; the weight calculation unit is used to obtain the weight of each road on the road network.

[0034] In another embodiment of the present invention, the present invention provides a smart home travel safety monitoring method, comprising the following steps:

[0035] Establish a road network, use monitoring devices on the road to collect the number of pedestrians on the road network and the time, use big data analysis to analyze the movement data of pedestrians on the road network, and obtain the relationship between pedestrians and time on each road in the road network;

[0036] The location and destination information of the elderly are obtained, and the probability of pedestrians passing each road on the road network is predicted when the elderly move on the road network. Combined with the lighting conditions on the road, the elderly are guided to their destination.

[0037] Using big data to analyze pedestrian movement data on the road network to obtain the relationship between pedestrians and time on each road in the road network also includes the following steps:

[0038] S10, taking the intersections on the road network as nodes and each road as an edge, obtains the time points t0 and t1 from the historical data of the i-th node. end The inflow data of the i-th node between them is anum, where t0 represents the time when the elderly start to move, t end represents the maximum time for the elderly to complete the movement; anum=(hn1+hn2+…+hnk) / k, where hn1, hn2,…, hnk represent the nodes at time points t0 and t end The historical personnel inflow data between, k is the number of historical personnel inflow data;

[0039] S11, obtain the historical movement path data of pedestrians entering the road network from the i-th node, let path j Represents the jth moving path of a pedestrian entering the road network from the i-th node, and obtains the probability P of the j-th moving path appearing j , P j =n j / sum i , where n j Indicates that the pedestrian entering the road network from the i-th node follows the path path j Number of moves, sum i represents the total number of pedestrians entering the road network from the i-th node;

[0040] S12, based on the movement path data and inflow data of the pedestrian at the i-th node, obtain the pedestrian entering from the i-th node on the road network at time points t0 and t end The location information between the time points t0 and t1 is obtained based on the existing movement path data of pedestrians on the road network. end The location information between the two nodes; according to the location information between the pedestrians entering from all nodes at time points t0 and t end The location information between the pedestrians at time points t0 and t endThe location information between them is used to obtain the unmanned time of each road on the road network, and the weight of the road is obtained according to the unmanned time of the road.

[0041] According to the existing movement path data of pedestrians on the road network, the movement paths of pedestrians on the road network at time points t0 and t end The location information between them also includes the following steps:

[0042] Get the historical movement data of all pedestrians from the historical movement data on the road network, and let the uth historical movement path containing the existing movement path of the ith existing pedestrian on the road network be path u , calculate the u-th historical moving path u The probability P u , P u =n u / sum, where n u Indicates the u-th historical moving path path u The number of occurrences indicates the number of occurrences of all historical movement paths including the existing movement path of the first pedestrian on the road network;

[0043] According to the pedestrian's moving speed data and moving path data, the number of pedestrians at time points t0 and t end The location information between them is based on the moving path u The probability P u , get the I-th pedestrian appearing on the moving path path u The probability of the road on u Based on the pedestrian's movement speed data, the relationship between the status of the first pedestrian on the road network and time is obtained; the relationship between the position of the first pedestrian on all movement paths and the movement time is obtained. If there are intersecting roads on different movement paths of the first pedestrian, and the time when the first pedestrian arrives at the intersecting road along different movement paths intersects, the probabilities of the different movement paths are added and the minimum value is taken between the sum and one to obtain the relationship between the status of the first pedestrian on the road network and time;

[0044] Get all existing pedestrians at time points t0 and t end According to the situation where there are intersecting roads on the different moving paths of the first pedestrian, the same steps are performed to obtain the relationship between the status of all existing pedestrians on the road network and time.

[0045] Pedestrians on the road can play a role in providing first aid and alerting the elderly when they fall. Therefore, the weight of the road is obtained according to the number of pedestrians on the road. The more time there are people on the road, the more likely it is that the elderly will be discovered when they fall, which will speed up the time it takes for the elderly to be rescued, thereby reducing the danger brought by the elderly falling.

[0046] For example, based on the existing movement path data of the first pedestrian on the road network, there are two possible movement paths for the first pedestrian, passing through nodes A1, A2, A3, B and B1, B2, A3, B. The probability of these two movement paths is the same, so when the first pedestrian moves, the impact of each movement path on the weight is 0.5; when the first pedestrian moves along the two paths, the time to arrive at the node is t11, t12, t13, t14 and t21, t22, t23, t24;

[0047] For the first moving path, within time (t0, t11), the probability of the first pedestrian appearing on the road between the starting point and node A1 is 50%, because the probability of the first moving path appearing is 50%; within time (t11, t12), the first pedestrian appears on the road between node A1 and node A2; then add the second moving path, because the two moving paths have an intersecting road, that is, the road between node A3 and node B, if there is no temporal intersection when moving according to these two moving paths, then according to the first moving path, within time (t13, t14), the first pedestrian has a 50% probability of appearing on the road between node A3 and node B, and according to the second moving path, within time (t23, t24), the first pedestrian has a 50% probability of appearing The probability of the first pedestrian appearing on the road between node A3 and node B is 50%; if there is a time intersection, for example, (t23, t24) is between (t13, t14), then within the time (t13, t23), the probability of the first pedestrian appearing on the road between node A3 and node B is 50%, within the time (t23, t24), the probability of the first pedestrian appearing on the road between node A3 and node B is 50%+50%, and within the time (t24, t14), the probability of the first pedestrian appearing on the road between node A3 and node B is 50%; the existence of road intersection and time intersection indicates that the first pedestrian will appear on the intersecting road within the time of the intersection. The probability of a pedestrian appearing on the road is the probability that there is someone on the road, which can provide help for the elderly when they fall.

[0048] When there are multiple pedestrians in the road network at time point t0, other pedestrians can also be regarded as the Ith pedestrian, which is equivalent to the Ith pedestrian starting from different starting points along different movement paths. The sum of the probabilities of all movement paths appearing is the number of pedestrians in the road network at time point t0. The sum of the probabilities of the movement paths appearing is used to reflect the number of people in the road network. When there is only one pedestrian in the road network, the sum of the probabilities of the movement paths appearing is 1; when there are n pedestrians in the road network, the sum of the probabilities of the movement paths appearing is n; in this way, according to the Ith pedestrian's processing method for different movement paths, the same operation is performed on multiple pedestrians to obtain the relationship between the status of all existing pedestrians on the road network and time; according to the relationship between the status of all existing pedestrians on the road network and time, the probability of there being people on each road on the road network at each time point within the time (t0, tend) is obtained.

[0049] t end It can be set to a larger value to ensure that the time it takes for the elderly to reach their destination after path planning does not exceed t end , thus obtaining the weight of the elderly on the road network during their movement.

[0050] According to the movement path data and inflow data of the pedestrian at the i-th node, the pedestrian entering from the i-th node on the road network at time points t0 and t end The location information between them also includes the following steps:

[0051] Get time points t0 and t end The personnel inflow data of the i-th node between time points t0 and t end The personnel inflow data of the i-th node is used to obtain the average time interval avet of the personnel inflow of the i-th node, avet=anum / (t end -t0), according to the time interval avet, generate virtual models of m pedestrians and put them into the i-th node of the road network at time points t0+dt, t0+dt+avet, t0+dt+2×avet, ... t0+dt+m×avet. According to the analysis method of the i-th existing pedestrian on the road network, after the virtual models of m pedestrians are put into the i-th node of the road network, the relationship between the status of the virtual pedestrian at the i-th node on the road network and time is obtained. The relationship between the status of the virtual pedestrians at all nodes on the road network and time is obtained, and the relationship between the status of all virtual pedestrians on the road network and time is obtained. Where m is (t end -t0) is the rounding of the time interval avet, dt is the bias, and is set according to the time when no pedestrians flow into the i-th node at time point t0.

[0052] First, the bias dt is obtained based on the current time point t0 and the time from the i-th node to the last pedestrian inflow; then, at the time point t0+dt, the virtual model of the pedestrian invested in the i-th node is regarded as a pedestrian already in the road network. In this way, the virtual model of the pedestrian can be analyzed according to the i-th existing pedestrian on the road network, and the relationship between the status of all virtual pedestrians on the road network and time can be obtained; the difference is that there is no existing movement path for the virtual pedestrian, so the analysis method of the probability of the movement path is different from that of the i-th existing pedestrian on the road network. Here, the historical movement path data of all pedestrians flowing into the road network from the i-th node are selected for analysis to obtain the probability of each movement path of the pedestrian flowing into the road network from the i-th node. The remaining part is the same as the analysis method of the i-th existing pedestrian on the road network.

[0053] According to the pedestrians entering from all nodes at time points t0 and t end The location information between the pedestrians at time points t0 and t end The unmanned time of each road on the road network is obtained by using the location information between the two roads, and the following steps are also included:

[0054] Obtain the relationship between all existing pedestrian conditions and time on the road network and the relationship between all virtual pedestrian conditions and time on the road network.

[0055] Obtaining the weight of a road based on its unoccupied time also includes the following steps:

[0056] Get time points t0 and t end The unmanned time period of the i-th road on the road network, let ts and te represent the starting point and end point of an unmanned time period, and calculate the pedestrian weight x3 brought to the road by an unmanned time period. , add up the pedestrian weights brought to the road by all unmanned time periods to obtain the pedestrian weight of the i-th road.

[0057] The weight of a road is positively correlated with the unmanned time period. During the unmanned time period, after an elderly person falls, they have to wait until the next manned time period to be discovered by pedestrians and rescued. However, the elderly person's falls on the road are random and they are likely to fall at any time point. The risk of falling is related to time. For example, if an elderly person falls at time point ts, they need to wait for te-ts time before the next pedestrian enters the road. However, if an elderly person falls at time point ts+0.5 (te-ts), they only need to wait 0.5 (te-ts) time before the next pedestrian enters the road. The risk of falling for the elderly is positively correlated with the time of falling. Therefore, the weight of the road is determined according to the time when no one is on the road after the elderly person falls.

[0058] The lighting conditions of the road are determined by the following steps:

[0059] The monitoring device on the road network obtains the image data of each road and converts the image data of each road into the linear RGB color space. For each pixel of the image data of each road, the brightness value is calculated, Y=0.2126R+0.7152G+0.0722B, where R is the red channel value of the image data, G is the green channel value of the image data, B is the blue channel value of the image data, and Y is the brightness value of the image data. Let the baseline brightness value of the road be Y max , calculate the relative brightness X of the pixels in the image data, X=Y / Y max , calculate the average brightness AX of the image data based on the relative brightness, and obtain the brightness weight x2 of the road based on the average brightness AX.

[0060] As the vision and sensory functions of the elderly are weakened, it is difficult for them to see the road conditions clearly, and they are prone to misjudging or missing things. This puts forward requirements for the lighting conditions of the roads, and determines the weight of the roads according to the lighting conditions of the roads.

[0061] Taking into account the lighting conditions on the road, guiding the elderly to their destination also includes the following steps:

[0062] S1, using Dijkstra algorithm for path planning, storing the elderly starting node in set U, and storing the remaining nodes except the elderly starting node in set V. If the elderly starting node is adjacent to other nodes in V, the weight between the elderly starting node and the adjacent nodes is the weight in the weighted graph; if they are not adjacent, the weight between the elderly starting node and the non-adjacent nodes is positive infinity;

[0063] S2, select a node with the smallest weight between it and the elderly person’s starting node from the set V and add it to the set U, and delete it from the set V;

[0064] S3, taking the node added to set U as the midpoint, modify the weights of each node in set V; if the weight of a node in set V passing through the midpoint to reach the elderly's starting node is smaller than the original weight, then modify the weight of set V. The modified weight is the weight of the elderly's starting node and the midpoint plus the weight of the node in V and the midpoint;

[0065] S4, repeat steps S2 and S3 until all nodes are included in set U.

[0066] The weight of the weighted graph is calculated using the following formula:

[0067] Z=W1×x1+W2×x2+W3×x3, where W1, W2 and W3 represent connection coefficients, x1 represents the length of the road; x1 and x2 are fixed values ​​that do not change with time, and x3 is a variable value that changes with time t0.

[0068] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. Intelligent home travel safety monitoring system, characterized by: It includes a monitoring module, a data storage module, a data analysis module, a path planning module and a guidance module; the monitoring module is used to obtain image data on the road network; the data storage module is used to store the historical movement data of pedestrians on the road network; the data analysis module is used to analyze the relationship between pedestrians on the road network and time; The path planning module plans a path for the elderly based on the relationship between pedestrians and time on the road network; The guiding module is used to guide the elderly from a starting point to a destination.

2. The intelligent home travel safety monitoring system according to claim 1, characterized in that: The path planning module further includes a network construction unit, a path planning unit, and a storage unit; the network construction unit is used to generate a weighted graph of a road network, including intersections, roads, and starting and destination information of the elderly in the road network; The path planning unit uses the Dijkstra algorithm to plan paths for the elderly based on the weighted graph of the road network; the storage unit is used to store the weighted graph information of the road network.

3. The intelligent home travel safety monitoring system according to claim 2, characterized in that: The path planning unit uses the Dijkstra algorithm to perform path planning, which specifically includes the following steps: S1, store the elderly starting node in the set U, and store the remaining nodes except the elderly starting node in the set V. If the elderly starting node is adjacent to other nodes in V, the weight between the elderly starting node and the adjacent nodes is the weight in the weighted graph; if they are not adjacent, the weight between the elderly starting node and the non-adjacent nodes is positive infinity; S2, select a node with the smallest weight between it and the elderly person’s starting node from the set V and add it to the set U, and delete it from the set V; S3, taking the node added to set U as the midpoint, modify the weights of each node in set V; if the weight of a node in set V passing through the midpoint to reach the elderly's starting node is smaller than the original weight, then modify the weight of set V. The modified weight is the weight of the elderly's starting node and the midpoint plus the weight of the node in V and the midpoint; S4, repeat steps S2 and S3 until all nodes are included in set U.

4. The intelligent home travel safety monitoring system according to claim 3 is characterized in that: The data analysis module uses big data to analyze the movement data of pedestrians on the road network to obtain the relationship between the number of pedestrians on each road in the road network and time; predicts the probability of pedestrians passing each road on the road network when the elderly person moves on the road network based on the starting point information and destination information; and uses big data to analyze the movement data of pedestrians on the road network to obtain the relationship between the number of pedestrians on each road in the road network and time, further comprising the following steps: S10, taking the intersections on the road network as nodes and each road as an edge, obtains the time points t0 and t1 from the historical data of the i-th node. end The inflow data of the i-th node between them is anum, where t0 represents the time when the elderly start to move, t end represents the maximum time for the elderly to complete the movement; anum=(hn1+hn2+…+hnk) / k, where hn1, hn2,…, hnk represent the nodes at time points t0 and t end The historical personnel inflow data between, k is the number of historical personnel inflow data; S11, obtain the historical movement path data of pedestrians entering the road network from the i-th node, let path j Represents the jth moving path of a pedestrian entering the road network from the i-th node, and obtains the probability P of the j-th moving path appearing j , P j =n j / sum i , where n j Indicates that the pedestrian entering the road network from the i-th node follows the path path j Number of moves, sum i represents the total number of pedestrians entering the road network from the i-th node; S12, based on the movement path data and inflow data of the pedestrian at the i-th node, obtain the pedestrian entering from the i-th node on the road network at time points t0 and t end The location information between the time points t0 and t1 is obtained based on the existing movement path data of pedestrians on the road network. end The location information between the two nodes; according to the location information between the pedestrians entering from all nodes at time points t0 and t end The location information between the pedestrians at time points t0 and t end The location information between them is used to obtain the unmanned time of each road on the road network, and the weight of the road is obtained according to the unmanned time of the road.

5. The intelligent home travel safety monitoring system according to claim 4 is characterized in that: The data analysis module further includes a space conversion unit, a first analysis unit, a second analysis unit, and a weight calculation unit; the space conversion unit is used to convert the image data of the road into a color space to obtain brightness information of the road; the first analysis unit is used to analyze the relationship between the conditions of all existing pedestrians on the road network and time; the second analysis unit is used to generate a virtual model of the pedestrian and obtain the relationship between the conditions of the virtual pedestrians on the road network and time; The weight calculation unit is used to obtain the weight of each road on the road network.

6. The intelligent home travel safety monitoring system according to claim 5, characterized in that: The space conversion unit obtains the image data of each road and converts the image data of each road into a linear RGB color space; for each pixel of the image data of each road, the brightness value is calculated, Y=0.2126R+0.7152G+0.0722B, where R is the red channel value of the image data, G is the green channel value of the image data, B is the blue channel value of the image data, and Y is the brightness value of the image data; let the baseline brightness value of the road be Y max , calculate the relative brightness X of the pixels in the image data, X=Y / Y max , calculate the average brightness AX of the image data based on the relative brightness, and obtain the brightness weight x2 of the road based on the average brightness AX.

7. The intelligent home travel safety monitoring system according to claim 6, characterized in that: The first analysis unit obtains the number of pedestrians on the road network at time points t0 and t1 based on the existing movement path data of pedestrians on the road network. end The relationship between the location information and all existing pedestrian conditions and time on the road network includes the following steps: Get the historical movement data of all pedestrians from the historical movement data on the road network, and let the uth historical movement path containing the existing movement path of the ith existing pedestrian on the road network be path u , calculate the u-th historical moving path u The probability P u , P u =n u / sum, where n u Indicates the u-th historical moving path path u The number of occurrences indicates the number of occurrences of all historical movement paths including the existing movement path of the first pedestrian on the road network; According to the pedestrian's moving speed data and moving path data, the number of pedestrians at time points t0 and t end The location information between them is based on the moving path u The probability P u , get the I-th pedestrian appearing on the moving path path u The probability of the road on u , according to the pedestrian's moving speed data, the relationship between the I-th existing pedestrian status and time on the road network is obtained; Obtain the relationship between the position and travel time of the first pedestrian on all movement paths. If there are intersecting roads on different movement paths of the first pedestrian, and the time when the first pedestrian arrives at the intersecting road along different movement paths overlaps, then add the probabilities of the different movement paths and take the minimum value between them and one to obtain the relationship between the status of the first pedestrian on the road network and time. Get all existing pedestrians at time points t0 and t end According to the situation where there are intersecting roads on the different moving paths of the first pedestrian, the same steps are performed to obtain the relationship between the status of all existing pedestrians on the road network and time.

8. The intelligent home and travel safety monitoring system according to claim 7, characterized in that: The second analysis unit obtains the pedestrians entering the road network from the i-th node at time points t0 and t end The relationship between the location information and the status and time of all virtual pedestrians on the road network is specifically comprised of the following steps: Get time points t0 and t end The personnel inflow data of the i-th node between time points t0 and t end The personnel inflow data of the i-th node is used to obtain the average time interval avet of the personnel inflow of the i-th node, avet=anum / (t end -t0), according to the time interval avet, generate virtual models of m pedestrians and put them into the i-th node of the road network at time points t0+dt, t0+dt+avet, t0+dt+2×avet, ... t0+dt+m×avet. According to the analysis method of the i-th existing pedestrian on the road network, after the virtual models of m pedestrians are put into the i-th node of the road network, the relationship between the status of the virtual pedestrian at the i-th node on the road network and time is obtained. The relationship between the status of the virtual pedestrians at all nodes on the road network and time is obtained, and the relationship between the status of all virtual pedestrians on the road network and time is obtained. Where m is (t end -t0) is the rounding of the time interval avet, dt is the bias, and is set according to the time when no pedestrians flow into the i-th node at time point t0.

9. The intelligent home and travel safety monitoring system according to claim 8, characterized in that: Obtaining the weight of a road based on its unoccupied time also includes the following steps: Get time points t0 and t end The unmanned time period of the i-th road on the road network, let ts and te represent the starting point and end point of an unmanned time period, and calculate the pedestrian weight x3 brought to the road by an unmanned time period. , add up the pedestrian weights brought to the road by all unmanned time periods to obtain the pedestrian weight of the i-th road.

10. The intelligent home and travel safety monitoring system according to claim 9, characterized in that: The weight calculation unit calculates the weight of the weighted graph using the following formula: Z=W1×x1+W2×x2+W3×x3, where Z represents the weight, W1, W2, and W3 represent the connection coefficients, and x1 represents the length of the road; x1 and x2 are fixed values ​​that do not change over time, and x3 is a variable value that changes with time t0.

Images