CN100507963C - Large range battlefield situation intelligent perception system and perception method - Google Patents

Abstract

The invention discloses a large-scale battlefield situation intelligent perception system, which includes a wireless sensor network and a terminal receiving device. The wireless sensor network is composed of self-organized wireless sensor nodes. Each node includes a processor module, a storage module, and a wireless communication module, and integrates a GPS module, an electronic compass module, a laser ranging module, and multiple video acquisition modules. The wireless sensor node performs motion detection, target identification and classification through the video acquisition module, and then performs target distance measurement, determines the target longitude and latitude information, and spreads in the network through multiple hops. The sensors work together to complete the perception of the battlefield situation. The terminal receiving device collects battlefield situation information in real time through a multi-hop wireless sensor network and converts it into a two-dimensional plane image display. The invention has the characteristics of large-scale multi-target detection, rapid deployment, real-time accurate information acquisition, strong anti-strike ability, support for multi-terminal access, strong scalability, etc., and has broad market prospects and applications in various military and civilian applications value.

Description

Large-scale battlefield situation intelligent perception system and perception method

technical field

The invention relates to a wireless sensor network intelligent perception system and method, in particular to a large-scale battlefield situation intelligent perception system and method that utilizes a distributed multi-video multi-hop wireless sensor network to realize multi-target detection, classification, positioning and tracking.

Background technique

In the battlefield environment, the grasp of the battlefield situation plays a very important role. If the information collection and processing are not timely or appropriate, the initiative may be lost and extremely heavy losses will be caused. Therefore, the battlefield situational awareness system is very necessary. In view of the urgency, suddenness, large-scale and vulnerability of the battlefield environment, the battlefield situational awareness system is required to be able to deploy quickly, conduct large-scale multi-target surveillance, and obtain accurate information. Real-time, at the same time, strong anti-strike capability and support for multi-terminal access are also necessary for the battlefield environment. This has become an urgent problem to be solved in the industry.

Contents of the invention

The purpose of the present invention is to address the deficiencies in the prior art, to provide a method for sensing the situation of a large-scale battlefield, mainly to solve the problem of using a multi-hop wireless sensor network for multi-target identification, classification and positioning in a large-scale battlefield environment, to achieve real-time Accurate information acquisition, grasp the battlefield situation.

For accomplishing the purpose of the present invention, the technical scheme that the present invention takes is:

A method of sensing a large-scale battlefield situation is characterized in that it comprises the following steps:

1) The wireless sensor node performs motion detection, target recognition and classification on the target through the video acquisition module;

2) The wireless sensor node measures the distance of the target through the GPS module, the electronic compass module and the laser ranging module, and obtains the position and time information of the target;

3) The wireless sensor node transmits the above information to the wireless sensor network through the wireless communication module;

4) After other wireless sensor nodes receive the information propagated on the Internet, repeat steps 1)-step 3), and carry out multi-hop propagation in the network;

5) The terminal receiving device collects the battlefield situation information sent by each wireless sensor node in real time from the wireless sensor network, and converts it into a two-dimensional plane image display, wherein:

Each sensor node maintains a static feature library and a space-time related target information library. The static feature library is the initial preset feature library. The classification and various characteristics of the target are set in the static feature library. The space-time related target information library is The target information base is dynamically generated, updated and maintained during the system operation, and the space-time related target information base includes the target information captured, identified and classified by the sensor node;

In step 1), the sensor node performs motion detection, target recognition, and target classification on the target captured by the video acquisition module according to the static feature library set;

In step 2), the sensor node performs positioning calculation on the target to obtain the position of the target, and the sensor node updates the space-time related target information database after finding the target;

In step 3), the sensor node sends the first type of information and the second type of information to the network, the first type of information includes the position of the target, the time when the target is at the target position, the identity and position of the node itself, and the description information of the target feature , the second type of information includes the position of the target, the time when the target is at the target position, the type and identification of the target;

In step 4), after receiving one type of information, other sensor nodes first need to update their own space-time related target information base according to this type of information, and send a message to update the target information base, and then repeat steps 1)- Step 3), after receiving the second-type information, other sensor nodes forward the second-type information directly.

Compared with the prior art, the beneficial effects of the present invention's large-scale battlefield situation intelligent perception system are:

The wireless sensor network in this system is constructed in the form of multi-hop self-organization, which is suitable for deployment in a large-scale environment. At the same time, because it can be deployed by airborne and other methods, the rapid deployment of the system can be realized. In addition, this system uses wireless sensors as equipment for information collection, processing and dissemination, which saves wiring overhead. At the same time, as an independent whole, the sensors are also very convenient to place, suitable for battlefields and various complex terrain environments. Based on this, each wireless sensor integrates multiple video acquisition modules, which increases the monitoring range and is suitable for simultaneous monitoring of multiple targets; moreover, each sensor integrates a GPS module, an electronic compass module and a wireless transmission module to obtain the location Information and orientation information, and exchange information through a wireless multi-hop network, so as to perform target identification, classification, positioning and tracking more quickly and accurately. In addition, this system adopts a distributed wireless sensor network, there is no central node, and the whole system can continue to work after several sensor nodes are damaged. Anti-strike capability and scalability. After each sensor obtains the video information, it calculates locally, and finally the information transmitted in the network only includes the target type, identification, target location and time, occupies less bandwidth, and has low requirements for terminal processing capabilities. Multiple terminals can simultaneously connect input, obtain target information and display two-dimensional trajectory. In addition, each wireless sensor node in this system calculates and transmits battlefield target information in real time, and the information provided by other sensors can be used between multiple sensors to quickly identify and confirm the target multiple times, which improves the real-time and accuracy of information . Finally, the system can realize the recognition and classification of various types of objects by setting the feature library, such as armored vehicles, tanks and individual soldiers, and has good scalability.

In short, the present invention has the characteristics of large-scale multi-target detection, rapid deployment, real-time accurate information acquisition, strong anti-strike capability, support for multi-terminal access, and strong scalability. It has broad market prospects and Value. In military terms, it can be used for battlefield situation awareness, target intrusion detection and anti-reconnaissance, etc. In civil terms, it can be used for real-time and dynamic emergency handling, such as disaster relief sites.

Description of drawings:

Fig. 1 is a wireless sensor node module diagram;

Figure 2 is a type of information format;

Fig. 3 is the second type information format;

Fig. 4 is a multi-wireless sensor cooperation flowchart;

Fig. 5 is a flow chart of wireless sensor node target identification, classification and information processing;

Fig. 6 is a flow chart of wireless sensor node receiving information processing;

Fig. 7 is a schematic diagram of the wide-range battlefield situation intelligent perception system of the present invention.

Detailed ways:

The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

The invention arranges several wireless sensor nodes within a certain range, self-organizes to build a multi-hop wireless sensor network, and then connects several terminal receiving devices to the wireless sensor network. As shown in Figure 1, each sensor node includes a processor module, a storage module, and a wireless communication module, and each sensor node is integrated with a GPS module, an electronic compass module, a laser ranging module, and multiple video acquisition modules. The sensor can obtain its position and direction information through its own functional modules. Among them: the processor module is used to process data; the storage module is used to store data; the wireless communication module is used to communicate with the terminal receiving device and other wireless sensor nodes in the wireless sensor network; the GPS module is used to obtain its own position and time data; the electronic compass module is used to obtain its own direction data; the video acquisition module is used to collect the image of the target; the laser ranging module is used to obtain the distance between the target and itself; the above modules are connected to the processor module, and the wireless sensor The nodes obtain their own position and direction data through the above functional modules and can locate the target; each wireless sensor node cooperates by receiving information sent by other wireless sensor nodes and sending information to the network to complete the awareness of the battlefield situation; The terminal receiving device is connected to the wireless sensor network, collects battlefield situation information in real time and converts it into a two-dimensional plane image display.

There are mainly two types of information in the system. One type of information is shown in Figure 2, including four parts: (1) location, describing the longitude and latitude position information of the target; (2) time, describing the moment when the target is at the target location; (3) source Sensor refers to the source node that sends this information, specifically including source node identification and source node location information; (4) target feature description, a specific description of the target, such as category, color, position, and speed. The second type of information is shown in Figure 3, including three parts: (1) target, specifically including the type and identification of the target; (2) location, describing the longitude and latitude position information of the target; (3) time, describing the target's location moment. Nodes use different mechanisms to process the construction and transmission of two types of information: when the node's space-time related target information database is updated, the corresponding type of information will be constructed and broadcast; while the second type of information will be periodically constructed and broadcast by the node. While constructing and sending its own information, a node also forwards information for other nodes to realize a wireless multi-hop network.

Each sensor node maintains two target information feature libraries: (1) Static feature library refers to the initial preset feature library, how many types of targets are divided into, and what are the characteristics of each type, all of which are set in the static feature library of. According to the set static feature library, each sensor performs motion detection, target recognition and target classification on the information captured by the video acquisition module, and then performs target positioning calculation to give the longitude and latitude position of the target. (2) Spatio-temporal related target information base is a target information base dynamically generated, updated and maintained during system operation, including the target information captured, identified and classified by the sensor node. After the node discovers the target, it first performs information processing and extraction according to the static feature library, and then adds the target information (classification, location information and target features) to the spatio-temporal related target information base. It will be jointly updated and maintained by this node and a class of information in the network, and sent to the network in the form of a class of information.

Please refer to Figure 4, the system perceives the situation of a large-scale battlefield through the following steps:

1) The wireless sensor node performs motion detection, target recognition and classification on the target through the video acquisition module;

2) The wireless sensor node measures the distance of the target through the GPS module, the electronic compass module and the laser ranging module, and obtains the position and time information of the target;

3) The wireless sensor node constructs an information packet with the above information, and spreads it to the wireless sensor network through the wireless communication module;

4) After other wireless sensor nodes receive the information propagated on the Internet, repeat steps 1)-step 3), and carry out multi-hop propagation in the network;

5) The terminal receiving device collects the battlefield situation information sent by each wireless sensor node in real time from the wireless sensor network, and converts it into a two-dimensional plane image display.

In the above method, the wireless sensor node performs motion detection, target recognition, and target classification on the information captured by the video acquisition module according to the set feature library, and then performs target positioning calculation, gives the longitude and latitude position of the target, and constructs an information packet And propagate through multiple hops in the network. The sensors work together to continuously exchange and send information to complete the perception of the battlefield situation. The terminal receiving device is connected to the wireless sensor network, collects battlefield situation information in real time and converts it into a two-dimensional plane image display.

Wherein: in step 1), the sensor node performs motion detection, target recognition, and target classification to the target captured by the video acquisition module according to the static feature library set; in step 2), the sensor node locates the target Calculate and obtain the position of the target, and the sensor node updates the space-time related target information database after finding the target; in step 3), the sensor node sends the first-class information and second-class information to the network. The process of identifying, classifying and information processing of targets by the wireless sensor nodes is shown in Fig. 5 . In Figure 5, node A represents the sending node, and node B represents the intermediate receiving node. Node A performs area detection through the video detection module, and extracts moving objects through motion detection technology (finding the moving area through changes between consecutive video frames). When a moving target is found, it will identify and classify the target based on the information of the static feature library, and then use the laser ranging module to measure the distance of the locked target, and calculate the target according to the position information provided by the GPS of this node and the angle information provided by the electronic compass Location. Then update the space-time related target information database, and construct a corresponding type of information to broadcast in the network. After node B receives a type of information sent by node A, it will activate the update signal of the space-time related target information database, and the update signal will prompt the node to quickly identify, classify and locate the target according to the content of the space-time related target information base. Then according to the node's self-discrimination results, operations such as target information correction, confirmation and update are performed on the content of the spatio-temporal related target information database. Finally, a corresponding type of information is constructed and broadcast in the network.

After receiving one type of information, the sensor node needs to update its own spatio-temporal related target information database. The process is shown in Figure 6. After the sensor node receives the information in the network, it first judges the type of information. If it is the second type of information, it judges whether the time difference is greater than the time threshold. If it is larger, it discards the information, otherwise it forwards the second type of information. If it is a type of information, then extract relevant information according to the format of a type of information, first determine whether the source sending node is itself, if so, discard the information, and further determine whether the time difference is greater than the time threshold, if it is greater than the time threshold, discard this information, otherwise further determine the source Whether the distance between the sending node and itself is greater than the threshold, and if so, discard it. If it is judged that this type of information is not timed out and is within its valid range, it will update its own space-time related target information database according to this information, and forward this information. Each sensor works together through two types of information and two types of information target information feature databases to complete the identification, classification, positioning and tracking of targets. Through wireless access to the network, the terminal collects the second type of information in real time and displays the two-dimensional target trajectory.

An embodiment of the present invention is described below.

Figure 7 shows a large-scale battlefield situation intelligent awareness system, including a wireless sensor network (wireless sensor nodes are self-organized, represented by S1 to S5 in this embodiment) and a terminal receiving device (in this embodiment, Z1 hint). The aforementioned wireless sensor nodes are always in an active state, and perform uninterrupted monitoring of the scene area to be monitored. As shown in Figure 7, a number of wireless sensor nodes are placed in a battlefield environment with a length and width of 3 kilometers to build a wireless sensor network. Each sensor node can obtain position information according to the GPS module and direction information according to the electronic compass.

At a certain moment, the targets T1 and T2 enter the monitoring range of S1. Taking T1 as an example, first S1 finds the moving target through motion detection, and performs target recognition and target classification on the target according to its static feature library, and obtains the target relative distance through the laser ranging module. The distance of the node, and then calculate the target latitude and longitude according to the GPS information of the node. After analysis, it was determined that the target was a tank with a label of 001, and its color and other related features were extracted at the same time. The first type of information (as shown in FIG. 2 ) and the second type of information (as shown in FIG. 3 ) are constructed from the above information. One type of information content is the target position and its time, the identification and position of the wireless sensor node, and other characteristics of the target, here is the color information, which is used to provide information for other sensor nodes. Then, the sensor node will add/update the target information (classification, location information and target features) to the space-time related target information database according to this information, and broadcast this type of information through the wireless communication module at the same time. The second type of information is the target location, time, target type and identification, which is used to provide information for the terminal and broadcast periodically through the wireless communication module.

It is known from Fig. 7 that wireless sensor nodes S2 and S3 are one-hop neighbors of S1. Taking S3 as an example below, when S3 receives a type of information sent by S1, it extracts relevant information according to the format of a type of information, first judges whether the source sending node is itself, and if so, discards the information, here is S1, and further judges Whether the time difference is greater than the time threshold, if greater, discard this information, otherwise further judge whether the distance between the source sending node and itself is greater than the threshold, if greater, discard. After judging, S3 finds that this type of information has not timed out and is within its valid range, so it updates its own space-time related target information database according to this information, forwards this information, and activates the update signal at the same time. The update signal prompts S3 to quickly identify, classify and locate targets according to the contents of the space-time related target information database. According to the target location information provided by S1, S3 can quickly find the target. Quickly identify, classify and locate according to the characteristics of the target, and then repeat the steps of S1 to update the information base and broadcast a type of information.

When S3 receives the second-type information sent by S1, it only needs to judge whether the time difference is greater than the time threshold, and if so, discard the information; otherwise, it forwards the second-type information. The terminal Z1 is connected to the wireless sensor network, receives the second type of information in real time, analyzes the content of the information and displays it as a two-dimensional target trajectory on the screen. For example, after receiving multiple second-type information in a short time interval, the time information is the first priority, and the distance between the source sensor and the target is the second priority for analysis and processing.

In summary, the present invention completes multi-target recognition, classification and positioning in a large-scale battlefield environment, multi-hop information dissemination through a wireless sensor network, and the terminal receiving device collects and processes information in real time, and analyzes it into a two-dimensional target trajectory.

The above-mentioned terminal receiving device may have various forms, including a specific receiving display terminal, a computer installed with receiving processing software, or a personal digital assistant.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications should also be considered within the protection scope of the present invention.

Claims (4)

1. the method for a perception large range battlefield situation is characterized in that it may further comprise the steps:

1) the wireless senser node carries out motion detection, Target Recognition and classification by video acquisition module to target;

2) the wireless senser node is found range to target by GPS module, electrical compass module, laser ranging module, draws the position and the temporal information of target;

3) that above information is propagated into wireless sensing by wireless communication module is online for the wireless senser node;

4) after other wireless senser node is received the information of online propagation, repeating step 1)-step 3), in net, carry out multi-hop and propagate;

5) battlefield situation information that sends from each wireless senser node of the online real-time collecting of wireless sensing of terminal receiving trap, and be converted into two-dimensional image and show, wherein:

All safeguard static nature storehouse and temporal and spatial correlations target information storehouse on each sensor node, the static nature storehouse is initial default feature database, the classification of target setting and all kinds of characteristics in the static nature storehouse, temporal and spatial correlations target information storehouse is the target information storehouse that dynamically generates, upgrades and safeguard in system's operational process, and the temporal and spatial correlations target information comprises the target information that this sensor node is caught, discerned and classifies in the storehouse;

In step 1), sensor node is according to the static nature storehouse of setting, to carrying out motion detection, Target Recognition, target classification by the video acquisition module captured object;

In step 2) in, sensor node positions calculating to target, draws the position of target, and sensor node is finding promptly to upgrade temporal and spatial correlations target information storehouse after the target;

In step 3), sensor node sends a category information and two category informations in net, a described category information comprise the position, target of target in the sign of moment of target location, node self and position, the target signature descriptor, described two category informations comprise that the position, target of target are in the moment of target location, the type and the sign of target;

In step 4), other sensor node is after receiving a category information, at first to upgrade the temporal and spatial correlations target information storehouse of self according to this category information, and the message in target information storehouse is upgraded in transmission, and then repeating step 1)-step 3), other sensor node is directly transmitted this two category information after receiving two category informations.

2. the method for perception large range battlefield situation as claimed in claim 1 is characterized in that:

The wireless senser node adopts broadcast mode multi-hop in net to propagate transmission information.

3. the method for perception large range battlefield situation as claimed in claim 1 is characterized in that:

Described step 4) is further divided into following substep:

4.1) after the wireless senser node receives the information of online propagation, judge the kind of information, if two category informations then change step 4.2), if a category information then changes step 4.3);

4.2) whether the judgement time difference greater than time threshold, as greater than then abandoning this information, otherwise transmits this two category information;

4.3) judge that the source sends whether node is self, then abandons this information in this way;

4.4) further whether the judgement time difference greater than time threshold, as greater than then abandoning this information;

4.5) further judge the source send node with self apart from whether greater than threshold value, as greater than then abandoning this information;

4.6) according to the temporal and spatial correlations target information storehouse of this information updating self;

4.7) this information is sent in net, activate update signal simultaneously, repeating step 1)-step 3).

4. as the method for the described perception large range battlefield situation of one of claim 1-3, it is characterized in that: a plurality of terminal receiving traps insert wireless sense network simultaneously, and real-time rendering target two dimension trajectory diagram also provides target identification, type, location and time information.

Images