Independent component analysis based automobile sound identification method
A technology of independent component analysis and sound recognition, which is used in traffic control systems, instruments, and traffic control systems of road vehicles.
Inactive Publication Date: 2009-07-01
BEIHANG UNIV
View PDF3 Cites 4 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In the recognition of vehicle sounds, due to the influence of the environment on the extracted sound, the speed of the vehicle, and the individual differences in the same type of vehicles, the establishment of a feature library for vehicle sounds is a problem of extracting effective features from a small sample training set.
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0050] The present invention will be further described in detail below in conjunction with the accompanying drawings.
[0051] The goal of the demonstration system is to classify roadside vehicles and distinguish between light vehicles and heavy vehicles.
[0052] The vehicle sound recognition method based on independent component analysis uses the frequency domain information of vehicle sound to carry out vehicle recognition and classification, including two processes of vehicle sound feature library training and vehicle sound feature recognition and classification. The adjusted spectrum vector is obtained by these two processes. For the common processing process, the processing process of this part is collectively referred to as the preprocessing process hereinafter.
[0053] 1. Pretreatment process
[0054] Step 1: Use an audio collection device to collect vehicle sound signals. Since 90% of the energy of the vehicle sound is concentrated below 4000HZ, the sampling freque...
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
Login to View More Abstract
The invention relates to a vehicle sound recognition method based on independent component analysis, which belongs to the technical field of audio processing and pattern recognition. The steps are: preprocessing various types of vehicle sounds to obtain the spectrum of various types of vehicle sounds, and cutting the spectrum according to the characteristics of vehicle sounds, and transforming the coefficients in the frequency domain to the logarithmic domain to improve vehicle sound recognition. Robustness; use independent component analysis to extract the characteristics of vehicle sound, which can adapt to the small sample training features of vehicle recognition; reconstruct the samples to be recognized on the feature space composed of independent components, and obtain the samples to be recognized and the center of the vehicle class Euclidean distance, and according to this Euclidean distance, the recognition samples are classified. The invention can quickly and efficiently recognize vehicle sounds, and is especially suitable for real-time computing environment. It can be applied to: detection of passing vehicles in the military; intelligent transportation systems in civilian use and other application scenarios.
Description
technical field [0001] The invention relates to a vehicle sound recognition method based on independent component analysis, which belongs to the technical field of audio processing and pattern recognition. Background technique [0002] In the field of pattern recognition, independent component analysis (Independent Component Analysis, ICA) and principal component analysis (Principal Component Analysis, PCA) have many similarities in processing methods and ideas. Both methods can realize the linear representation of observed data. Both can project data from a high-dimensional space to a more meaningful low-dimensional space to achieve the purpose of dimensionality reduction and redundancy reduction, and perform corresponding processing, such as data structure analysis, data feature extraction, etc. The decomposition based on the PCA principle can only ensure that the decomposed components are uncorrelated, and cannot guarantee that these components are independent of each oth...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G08G1/017
Inventor 吴威曹靖耿云霄周忠赵沁平
Owner BEIHANG UNIV