Sonar Frequency Line Pattern Recognition for Vehicle Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for classifying vehicles based on their operating noise fail to automatically recognize a complete frequency line pattern within an amplitude spectrum, missing essential information such as doublets and not being operator-independent, leading to inefficiencies and potential inaccuracies in classification.

Innovation Solution

A method that automatically identifies a complete frequency line pattern by sorting and ranking frequency lines based on amplitude, forming line pairs, and determining final frequency line sets, which are then used to derive characteristic parameters for classification, allowing for operator-independent noise classification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual input methods are used to enter frequency line pattern data, then the operator can control the input process, but a lot of time is lost in the classification and information may be overlooked or not entered with sufficient accuracy

Engineering Contradiction:
Improveaccuracy of frequency line pattern inputVSAvoidtime lost in classification
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system automatically performs frequency line detection, sorting, and pattern recognition without requiring manual operator intervention. The computer system processes the amplitude spectrum itself to identify and classify frequency lines, eliminating the need for operators to manually enter data while maintaining high accuracy through automated algorithms

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The manual mechanical process of operators using graphic input tools is replaced by an automated computer-based system that uses signal processing algorithms to detect and analyze frequency lines in the amplitude spectrum, substituting human manual operations with automated computational methods

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Device complexity

If only three frequency lines with the largest amplitude values are considered, then the determination process is simplified, but frequency lines with lower amplitude and individual lines within the frequency spectrum are not recorded

Engineering Contradiction:
Improvecomplexity of frequency line analysisVSAvoidfrequency line information
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The frequency spectrum is segmented into multiple groups based on amplitude thresholds. Instead of considering only the top three frequency lines, the system divides all detectable frequency lines into different amplitude ranges and processes each segment, ensuring that both high-amplitude and low-amplitude frequency lines are captured and analyzed

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs excessive analysis by examining all frequency lines above a certain threshold rather than limiting to just three lines. This partial or excessive action ensures that no potentially important frequency information is missed, even if it increases the amount of data processed

Inventive Principle:
Principle #16Partial or excessive action

3Ease of operation

If individual fundamental frequencies are determined based on three frequency lines, then the determination process is straightforward, but it is not possible to record a complete set of frequency lines or a complete frequency line pattern including doublets

Engineering Contradiction:
Improveease of frequency line determinationVSAvoidcompleteness of frequency line pattern
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

An intermediary sorting step is introduced between raw frequency line detection and final pattern recognition. Frequency lines are sorted by amplitude and processed through multiple analysis stages, allowing the system to build a complete frequency line pattern that includes fundamental frequencies, harmonics, and doublets while maintaining operational simplicity through structured processing

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

Frequency lines are preliminarily sorted and ranked by amplitude before the main pattern recognition process. This preliminary organization of data prepares the information in advance, making the subsequent pattern detection more accurate and complete without complicating the overall operation

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2883074B1Method and apparatus for determining a frequency line pattern within at least one amplitude spectrum
Publication Date: 2016.06.08 ATLAS ELEKTRONIK GMBH
  • EP2883074B1 patent drawingFigure 1
  • EP2883074B1 patent drawingFigure 2~3

AI summary

The invention relates to a method for determining a frequency line pattern within at least one amplitude spectrum that is produced from audible signals that have been radiated by at least one detected vehicle and received by means of a sonar installation. First of all, a predetermined number of frequency lines in the or each amplitude spectrum are selected and are sorted in order of decreasing amplitude magnitude and are selected using a threshold value that ensures that a predetermined suitable number is not exceeded. Following stipulation of a plurality of threshold values, frequency lines that have been selected taking account of at least one threshold value are compared with one another. The line pairs formed by the comparison are used to determine preliminary frequency line sets with an associated fundamental frequency that are rated on the basis of their identifiable structures. An exclusion method is used to form the final frequency line sets, taking account of a rating that is dependent on the sequence, the number, the density, the amplitude and the fundamental frequency of the frequency lines. It is then possible to find strong single lines that the spectrum may contain. For each detected vehicle, it is therefore possible to assign a target-oriented frequency line pattern that can be used to classify the vehicle. The invention also relates to an appropriate apparatus for carrying out the method according to the invention.