Unlock instant, AI-driven research and patent intelligence for your innovation.

A Target Recognition Method for Handling Fuzzy and High Conflict Information

A target recognition and fuzzy technology, which is applied in the field of target recognition dealing with fuzzy and high-conflict information, can solve problems such as inability to resolve conflict information, inaccurate target detection, and high information ambiguity, and achieve simple merging methods, reduced calculations, The effect of avoiding the possibility

Active Publication Date: 2019-11-08
UNIV OF SCI & TECH BEIJING
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Furthermore, affected by the size, sensor manufacturing process and complex interference factors in the real environment, the information collected by nodes in wireless sensor networks not only has high ambiguity but also often has high conflicts.
If the fuzzy and conflict information collected by each node is discarded, only using the precise information of the node will inevitably lead to inaccurate target detection, or even wrong results, so it is particularly important to effectively deal with fuzzy and high conflict information
Although there are related data fusion algorithms, some of them can’t actually solve conflicting information. Even if some algorithms can solve it, there are still common problems such as excessive calculation, unreasonable generation of focal elements and “explosion of focal elements”

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Target Recognition Method for Handling Fuzzy and High Conflict Information
  • A Target Recognition Method for Handling Fuzzy and High Conflict Information
  • A Target Recognition Method for Handling Fuzzy and High Conflict Information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] Such as figure 1 As shown, the target recognition method for processing fuzzy and high-conflict information provided by the embodiment of the present invention includes:

[0053] S101. Obtain multiple evidences of the detected target, where the multiple evidences are multiple evidences containing fuzzy and highly conflicting information;

[0054] S102. Determine a corresponding recognition frame according to the detected target, and obtain a single focal unit under the recognition frame according to the determined recognition frame, and the single focal unit is a fuzzy and high-conflict information unit with supporting information Jiao Yuan;

[0055] S103, according to the obtained single focal unit, merge the fuzzy and high-conflict information single focal unit with supporting information into a new focal unit through a new combination method, and obtain a shape power set composed of the new focal unit, wherein the new The combination method is to use the operator s...

Embodiment 2

[0105] This embodiment also provides a target recognition system for processing fuzzy and high-conflict information. The system is used to implement the target recognition method for processing fuzzy and high-conflict information described in the embodiment of the present invention. The system includes: A target detection node, a target information collection node, an information processing node, and a discrimination node; the target detection node, the target information collection node, an information processing node, and a discrimination node are all sensor nodes in a wireless sensor network; the target detection node, the target The information collection nodes can be the same sensor nodes or different sensor nodes.

[0106] The workflow of the system can include:

[0107] Track the moving target through the target detection node. If the detected target appears, notify each sensor node in the target information collection node to start collecting the target information of ...

Embodiment 3

[0110] In order to better understand the target recognition method for processing fuzzy and high-conflict information described in this embodiment of the present invention, combined with specific examples, the target recognition method for processing fuzzy and high-conflict information described in this embodiment and The system is described in detail:

[0111] Such as image 3 As shown, the current application scenario is the field of vehicle recognition in the traffic system. Some magnetic sensor nodes and image sensor nodes are deployed on both sides of the road respectively. These nodes can be divided into clusters at intervals, and each group contains magnetic sensor nodes and image sensor nodes. These magnetic sensor nodes and image sensor nodes can communicate with each other and are inexpensive to form a heterogeneous sensor network; Among them, the magnetic sensor nodes can act as target detection nodes to detect the presence of the detected target, and after detect...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a target recognition method for processing fuzzy and high-conflict information, which can reduce the calculation amount and improve the accuracy of the detected target classification. The method includes: according to the obtained single focal elements under the identification framework, combining the fuzzy and high-conflict information single focal elements with supporting information into new focal elements through a new combination method, and obtaining the form power composed of the new focal elements set, wherein, the new combination method is to use the operator symbol ◇ to combine; according to the information supported by each evidence and the new focal element in the form power set and the basic confidence assignment function, using the information synthesis rules based on logical operations, the The plurality of evidences containing fuzzy and high-conflict information are fused to obtain a preliminary synthesis result for target identification; according to the identification framework and the preliminary synthesis result, the detected target is identified according to a preset decision rule. The invention relates to the field of object classification in wireless sensor networks.

Description

technical field [0001] The invention relates to the field of target classification in wireless sensor networks, in particular to a target recognition method for processing fuzzy and high-conflict information. Background technique [0002] In wireless sensor networks, it is often necessary to detect and classify objects through sensors. The existing target classification methods include decision tree classification method, Bayesian classification method, support vector machine classification method, backpropagation classification method and so on. However, these classification methods are mostly used in fields such as machine learning and data mining that need to process massive data, which not only requires high data volume, but also has high requirements for the storage and computing capabilities of machines. On the contrary, in wireless sensor network applications, due to the limited energy available to sensor nodes, resources such as internal storage capacity, data proce...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62
CPCG06F18/24
Inventor 王睿蔡仕娇
Owner UNIV OF SCI & TECH BEIJING