Sound Classification Method Research Articles

Classification of heart sounds based on quality assessment and wavelet scattering transform

Automatic classification of heart sound plays an important role in the diagnosis of cardiovascular diseases. In this study, a heart sound sample classification method based on quality assessment and wavelet scattering transform was proposed. First, the ratio of zero crossings (RZC) and the root mean square of successive differences (RMSSD) were used for assessing the quality of heart sound signal. The first signal segment conforming to the threshold standard was selected as the current sample for the continuous heart sound signal. Using the wavelet scattering transform, the wavelet scattering coefficients were expanded according to the wavelet scale dimension, to obtain the features. Support vector machine (SVM) was used for classification, and the classification results for the samples were obtained using the wavelet scale dimension voting approach. The effects of RZC and RMSSD on the results are discussed in detail. On the database of PhysioNet Computing in Cardiology Challenge 2016 (CinC 2016), the proposed method yields 92.23% accuracy (Acc), 96.62% sensitivity (Se), 90.65% specificity (Sp), and 93.64% measure of accuracy (Macc). The results show that the proposed method can effectively classify normal and abnormal heart sound samples with high accuracy.

Abnormal heart sound classification using phonocardiography signals

An intelligent support system is needed to assist in the identification of abnormalities of a human heart. The integration of signal processing with machine learning techniques is a new research trend in the studies of heart sound analysis. This paper proposes a heart sound feature dimension reduction and classification methods using supervised machine learning algorithms, by utilising the first (S1) and the second (S2) heart sounds, produced due to vibrations during the closure of heart valves. The features of S1 and S2 heart sounds are extracted in both time and frequency domains. Time domain features are based on S1 and S2 sound distance, amplitude, sound peak area, sound peak cycle duration and intensity, whilst 20 Mel Frequency Cepstral Coefficients (MFCCs) filter-bank energy for 12 coefficients represent the frequency domain features. Statistical values of the selected features are further used to increase the number of heart sound features. Due to the size of the extracted features, Linear Discriminant Analysis (LDA) dimensionality reduction technique has been used to select the best features for normal and abnormal heart sound classification using an Artificial Neural Network (ANN) model. It has been shown that the proposed LDA/ANN heart sound classification model achieved 90%, 83.33%, and 93.33% classification accuracies using the time domain, frequency domain and combined time-frequency domain features, respectively. The results using the proposed method are significantly better than previous classification methods by other researchers, with minimal complexity. This work provides a step forward in providing clinical informatics tool to assist clinician in providing early detections of abnormal heart conditions.

Application of substitution box of present cipher for automated detection of snoring sounds

Background and purposeSnoring is one of the sleep disorders, and snoring sounds have been used to diagnose many sleep-related diseases. However, the snoring sound classification is done manually which is time-consuming and prone to human errors. An automated snoring sound classification model is proposed to overcome these problems. Material and methodThis work proposes an automated snoring sound classification method using three new methods. These methods are maximum absolute pooling (MAP), the nonlinear present pattern, and two-layered neighborhood component analysis, and iterative neighborhood component analysis (NCAINCA) selector. Using these methods, a new snoring sound classification (SSC) model is presented. The MAP decomposition model is applied to snoring sounds to extract both low and high-level features. The presented model aims to attain high performance for SSC problem. The developed present pattern (Present-Pat) uses substitution box (SBox) and statistical feature generator. By deploying these feature generators, both textural and statistical features are generated. NCAINCA chooses the most informative/valuable features, and these selected features are fed to k-nearest neighbor (kNN) classifier with leave-one-out cross-validation (LOOCV). The Present-Pat based SSC system is developed using Munich-Passau Snore Sound Corpus (MPSSC) dataset comprising of four categories. ResultsOur model reached an accuracy and unweighted average recall (UAR) of 97.10 % and 97.60 %, respectively, using LOOCV. Moreover, a nocturnal sound dataset is used to show the universal success of the presented model. Our model attained an accuracy of 98.14 % using the used nocturnal sound dataset. ConclusionsOur developed classification model is ready to be tested with more data and can be used by sleep specialists to diagnose the sleep disorders based on snoring sounds.

Short-term feeding behaviour sound classification method for sheep using LSTM networks

A deep learning approach using long-short term memory (LSTM) networks was implemented in this study to classify the sound of short-term feeding behaviour of sheep, including biting, chewing, bolus regurgitation, and rumination chewing. The original acoustic signal was split into sound episodes using an endpoint detection method, where the thresholds of short-term energy and average zero-crossing rate were utilized. A discrete wavelet transform (DWT), Mel-frequency cepstral, and principal-component analysis (PCA) were integrated to extract the dimensionally reduced DWT based Mel-frequency cepstral coefficients (denoted by PW_MFCC) for each sound episode. Then, LSTM networks were employed to train classifiers for sound episode category classification. The performances of the LSTM classifiers with original Mel-frequency cepstral coefficients (MFCC), DWT based MFCC (denoted by W_MFCC), and PW_MFCC as the input feature coefficients were compared. Comparison results demonstrated that the introduction of DWT improved the classifier performance effectively, and PCA reduced the computational overhead without degrading classifier performance. The overall accuracy and comprehensive F1-score of the PW_MFCC based LSTM classifier were 94.97% and 97.41%, respectively. The classifier established in this study provided a foundation for an automatic identification system for sick sheep with abnormal feeding and rumination behaviour pattern. Keywords: sheep behaviour, short-term feeding behaviour, acoustic analysis, Mel-frequency cepstral coefficients, long-short term memory networks DOI: 10.25165/j.ijabe.20211402.6081 Citation: Duan G H, Zhang S F, Lu M Z, Okinda C, Shen M X, Norton T. Short-term feeding behaviour sound classification method for sheep using LSTM networks. Int J Agric & Biol Eng, 2021; 14(2): 43–54.

Novel three kernelled binary pattern feature extractor based automated PCG sound classification method

BackgroundHeart valve diseases are commonly seen ailments, and many people suffer from these diseases. Therefore, early diagnosis and accurate treatment are crucial for these disorders. This research aims to diagnose heart valve diseases automatically by employing a new stable feature generation method. Materials and methodThis research presents a stable feature generator-based automated heart diseases diagnosis model. This model uses three primary sections. They are stable feature generation using the improved one-dimensional binary pattern (IBP), selecting the most discriminative feature with neighborhood component analysis (NCA), and classification employing the conventional classifiers. IBP uses three kernels, and they are named signum, left signed, and right signed kernels. By applying these kernels, 768 features are generated. NCA aims to choose the most discriminative ones, and 64 features are chosen to employ NCA. The k nearest neighbor (kNN) and support vector machine (SVM) classifier are employed in the classification phase. Open access (public published) Phonocardiogram signal (PCG) sound dataset is used to calculate this model's measurements. This dataset contains 1000 PCGs with five categories. ResultsThe presented IBP and NCA-based heart valve disorders classification model tested using kNN and SVM classifier and attained 99.5% and 98.30% accuracies, respectively. ConclusionsPer the results, the presented IBP and NCA-based PCG sound classification is a successful method. Moreover, this model is basic and high accurate. Therefore, it is ready for the development of real-time implementations.

A heart sound classification method based on joint decision of extreme gradient boosting and deep neural network

Heart sound is one of the common medical signals for diagnosing cardiovascular diseases. This paper studies the binary classification between normal or abnormal heart sounds, and proposes a heart sound classification algorithm based on the joint decision of extreme gradient boosting (XGBoost) and deep neural network, achieving a further improvement in feature extraction and model accuracy. First, the preprocessed heart sound recordings are segmented into four status, and five categories of features are extracted from the signals based on segmentation. The first four categories of features are sieved through recursive feature elimination, which is used as the input of the XGBoost classifier. The last category is the Mel-frequency cepstral coefficient (MFCC), which is used as the input of long short-term memory network (LSTM). Considering the imbalance of the data set, these two classifiers are both improved with weights. Finally, the heterogeneous integrated decision method is adopted to obtain the prediction. The algorithm was applied to the open heart sound database of the PhysioNet Computing in Cardiology(CINC) Challenge in 2016 on the PhysioNet website, to test the sensitivity, specificity, modified accuracy and F score. The results were 93%, 89.4%, 91.2% and 91.3% respectively. Compared with the results of machine learning, convolutional neural networks (CNN) and other methods used by other researchers, the accuracy and sensibility have been obviously improved, which proves that the method in this paper could effectively improve the accuracy of heart sound signal classification, and has great potential in the clinical auxiliary diagnosis application of some cardiovascular diseases.

Multileveled ternary pattern and iterative ReliefF based bird sound classification

Birds may need to be identified for purposes such as environmental monitoring, follow-up, and species detection in the ecological area. Automatic sound classifiers have been used to perform species detection. Many methods have been presented in the literature to classify bird sounds with high accuracy. Nowadays, deep learning models have been used to classify data with high classification accuracy. However, these networks have high computational complexity. To obtain a highly accurate and lightweight classification model, a new multileveled and handcrafted features based machine learning model is presented. The presented automated bird sound classification model uses the multileveled ternary pattern (TP) feature generation, feature selection, and classification phases. The multileveled feature generation network can reach high classification accuracies since they generate high-level, low-level, and mid-level features. To construct levels, discrete wavelet transform (DWT) is employed to use the effectiveness of the DWT in bird sound classification. An improved version of the ReliefF, which is iterative ReliefF (IRF), is considered as feature selector. IRF selects the most informative features automatically, and these features are operated on linear discriminant (LD), k nearest neighbor (kNN), bagged tree (BT), and support vector machine (SVM) classifiers to calculate results of variable classifiers. The proposed multilevel TP and IRF based bird sound classification method reached 96.67% accuracy by using SVM on the 18 classes bird sound dataset.

Heart Sounds Classification Based on Feature Fusion Using Lightweight Neural Networks

Heart sounds are an important basis for evaluating heart disease. In recent years, auxiliary diagnosis technology of cardiovascular disease based on the detection of heart sound signals has become a research hotspot. A lightweight automatic heart sound classification method is presented in this study. It only needs to perform simple preprocessing on the heart sound data, and a neural network model is constructed to automatically extract the time–frequency features of heart sound data: where the convolution module extracts the frequency-domain characteristics, the loop module extracts the time-domain characteristics, and the features are then fused in series and parallel. Finally, the classification and recognition of heart sounds are realized based on the fusion features. At the same time, methods, such as group convolution, global average pooling, and gated loop mechanisms, are used to reduce the parameters and training time of the neural network. The experimental data from the PhysioNet database are used for testing. The recognition accuracy of the series feature fusion heart sound classification model is 95.00%, and the model size is 0.6 MB. In addition, the parallel feature fusion heart sound classification model is 95.50%, and the model size is 1.36 MB. Therefore, the lightweight heart sound classification model designed in this study contains fewer parameters and higher recognition accuracy. This makes it suitable for deployment in embedded devices. It also has important research significance for the development of portable heart sound detection equipment.

Short-term feeding behaviour sound classification method for sheep using LSTM networks

A deep learning approach using long-short term memory (LSTM) networks was implemented in this study to classify the sound of short-term feeding behaviour of sheep, including biting, chewing, bolus regurgitation, and rumination chewing. The original acoustic signal was split into sound episodes using an endpoint detection method, where the thresholds of short-term energy and average zero-crossing rate were utilized. A discrete wavelet transform (DWT), Mel-frequency cepstral, and principal-component analysis (PCA) were integrated to extract the dimensionally reduced DWT based Mel-frequency cepstral coefficients (denoted by PW_MFCC) for each sound episode. Then, LSTM networks were employed to train classifiers for sound episode category classification. The performances of the LSTM classifiers with original Mel-frequency cepstral coefficients (MFCC), DWT based MFCC (denoted by W_MFCC), and PW_MFCC as the input feature coefficients were compared. Comparison results demonstrated that the introduction of DWT improved the classifier performance effectively, and PCA reduced the computational overhead without degrading classifier performance. The overall accuracy and comprehensive F1-score of the PW_MFCC based LSTM classifier were 94.97% and 97.41%, respectively. The classifier established in this study provided a foundation for an automatic identification system for sick sheep with abnormal feeding and rumination behaviour pattern. Keywords: sheep behaviour, short-term feeding behaviour, acoustic analysis, Mel-frequency cepstral coefficients, long-short term memory networks DOI: 10.25165/j.ijabe.20211402.6081 Citation: Duan G H, Zhang S F, Lu M Z, Okinda C, Shen M X, Norton T. Short-term feeding behaviour sound classification method for sheep using LSTM networks. Int J Agric & Biol Eng, 2021; 14(2): 43–54.

An automated snoring sound classification method based on local dual octal pattern and iterative hybrid feature selector

In this research, a novel snoring sound classification (SSC) method is presented by proposing a new feature generation function to yield a high classification rate. The proposed feature extractor is named as Local Dual Octal Pattern (LDOP). A novel LDOP based SSC method is presented to solve the low success rate problems for Munich-Passau Snore Sound Corpus (MPSSC) dataset. Multilevel discrete wavelet transform (DWT) decomposition and the LDOP based feature generation, informative features selection with ReliefF and iterative neighborhood component analysis (RFINCA), and classification using k nearest neighbors (kNN) are fundamental phases of the proposed SSC method. Seven leveled DWT transform, and LDOP are used together to generate low, medium, and high levels features. This feature generation network extracts 4096 features in total. RFINCA selects 95 the most discriminative and informative ones of these 4096 features. In the classification phase, kNN with leave one out cross-validation (LOOCV) is used. 95.53% classification accuracy and 94.65% unweighted average recall (UAR) have been achieved using this method. The proposed LDOP based SSC method reaches 22% better result than the best of the other state-of-the-art machine learning and deep learning-based methods. These results clearly denote the success of the proposed SSC method.

FusedTSNet: An automated nocturnal sleep sound classification method based on a fused textural and statistical feature generation network

Nowadays, many people have suffered from sleep disorders. These diseases affect daily life and can disrupt sanity. Sleep disorders/diseases can be diagnosed by using nocturnal sleep sounds. This work presents an automated nocturnal sound classification method. The proposed nocturnal sound classification method can be used in the automated sleep disease diagnosis process. To propose a highly accurate and cognitive method, a fused feature generation network is proposed. The proposed fused feature generation network extracts both textural features and statistical features together. Therefore, this method is called as fused textural and statistical feature generation network (FusedTSNet). One-dimensional discrete wavelet transform (DWT) is employed to create levels and 7-leveled DWT is applied to nocturnal sounds. Here, DWT is utilized as a pooling/decomposition method to create a multileveled feature generation network. By using the ReliefF iterative neighborhood component analysis (RFINCA), the most valuable features are selected. To demonstrate the success of the FusedTSNet and RFINCA based nocturnal sound classification method, conventional classifiers are used. The proposed FusedTSNet and RFINCA based nocturnal sound classification method were tested on a collected nocturnal sound dataset. This dataset has 700 sounds in 7 classes. Our method achieved a 98.0% classification rate on this dataset. This work clearly indicates that the automated sleep behavior detection can be developed and the success of the proposed FusedTSNet and RFINCA based sound classification method is obviously shown.

A novel spiral pattern and 2D M4 pooling based environmental sound classification method

One of the crucial problems of the signal processing, digital forensics and machine learning is the environmental sound classification (ESC). Several ESC methods have been presented to obtain highly accurate model. In this work, a novel multileveled ESC method is presented. The presented ESC method uses two novel algorithms namely Spiral Pattern and two dimensional maximum, minimum, median and mean (2D-M4) pooling. By using these methods (Spiral Pattern and 2D-M4 pooling), 9 level feature generation approach is presented. Since the proposed Spiral Pattern has nine arrows, it extracts 9 and 18 bits using signum and ternary functions respectively. As a result, 1536 features are extracted in each level and totally 15,360 features are generated using from 0th to 9th levels. In order to select the discriminative features, neighbourhood component analysis (NCA) is used and 700 most distinctive features are selected. In the classification phase, deep neural network is trained and tested with the ESC-10 and ESC-50 datasets. 98.75% and 85.75% average classification accuracies were achieved with 10-folds cross validation for ESC-10 and ESC-50 datasets respectively. The experimental results reveal that the proposed Spiral Pattern and 2D-M4 pooling based ESC method is superior than the human auditory system (HAS) for environmental sound classification.

A novel ternary and signum kernelled linear hexadecimal pattern and hybrid feature selection based environmental sound classification method

Environmental sound classification (ESC) is one of the fundamental study areas for digital forensics and machine learning. A novel textural feature extractor which is ternary and signum kernelled linear hexadecimal pattern (TSK-LHP) is presented as feature extractor. Multileveled feature extraction method is used and levels are created by discrete wavelet transform (DWT). TSK-LHP generates features from each level. The most distinctive ones are selected by using hybrid feature selector. This hybrid feature selector uses neighborhood component analysis (NCA) and principle component analysis (PCA) together. Therefore, it is called as NPCA. A novel ESC dataset was collected for testing and there are 1211 sounds with 25 classes in this dataset. The proposed method is tested by using four shallow classifiers which are decision tree (DT), linear discriminant (LD), support vector machine (SVM), k nearest neighbor (kNN) and bagged tree (BT). Our proposed method achieved 99.83%, 100.0%, 99.17%, 93.64% and 98.35% classification accuracies by using these classifiers respectively.

Heart sound classification based on improved MFCC features and convolutional recurrent neural networks

Heart sound classification plays a vital role in the early detection of cardiovascular disorders, especially for small primary health care clinics. Despite that much progress has been made for heart sound classification in recent years, most of them are based on conventional segmented features and shallow structure based classifiers. These conventional acoustic representation and classification methods may be insufficient in characterizing heart sound, and generally suffer from a degraded performance due to the complicated and changeable cardiac acoustic environment. In this paper, we propose a new heart sound classification method based on improved Mel-frequency cepstrum coefficient (MFCC) features and convolutional recurrent neural networks. The Mel-frequency cepstrums are firstly calculated without dividing the heart sound signal. A new improved feature extraction scheme based on MFCC is proposed to elaborate the dynamic characteristics among consecutive heart sound signals. Finally, the MFCC-based features are fed to a deep convolutional and recurrent neural network (CRNN) for feature learning and later classification task. The proposed deep learning framework can take advantage of the encoded local characteristics extracted from the convolutional neural network (CNN) and the long-term dependencies captured by the recurrent neural network (RNN). Comprehensive studies on the performance of different network parameters and different network connection strategies are presented in this paper. Performance comparisons with state-of-the-art algorithms are given for discussions. Experiments show that, for the two-class classification problem (pathological or non-pathological), a classification accuracy of 98% has been achieved on the 2016 PhysioNet/CinC Challenge database.

An automated environmental sound classification methods based on statistical and textural feature

Determining the location from environmental sounds is crucial for digital forensics. Therefore, it is possible to predict about the sounds obtained using the automatic environmental sound classification (ESC) method. In this study, a method has been proposed for classifying environmental sounds. The main objective of this paper is to present a high accurate stable feature extraction based ESC method. This method consists of 3 fundamental stages and these are feature generation, selecting feature and classification stages. One dimensional local binary pattern (1D-LBP), One dimensional ternary pattern (1D-TP) and statistical feature generation methods are used for feature extraction. Neighborhood component analysis is used to select discriminative features and cubic (3rd Polynomial Order Kernel) support vector machine is used for classification. The proposed method is applied on ESC-10 dataset and the classification of the sounds on the dataset has been provided. 90.25% accuracy rate is obtained with the proposed method. A novel cognitive, high accurate and lightweight ESC method is presented in this work. In addition, comparison with previous studies is presented. The proposed method is outperformed.

Cardiac valves disorder classification based on active valves appearance periodic sequences tree of murmurs

Heart sound auscultation and interpretation is a primary diagnosis and cardiac physical examination method. Heart sound classification becomes a challenging issue in the present of complicated heart sound or multiple valvular heart diseases. Unlike the existence automatic heart sound classification techniques, a powerful heart sound classification method utilizing spatial-temporal information is proposed to determine the heart murmur type when multiple heart murmurs are occurred. First of all, normal and abnormal heart sounds are discriminated via power spectra density of heart sound feature extraction along with Support Vector Machine classifier. Spatial-temporal information of heart valves are extracted by microphone array recording and recursively sound source localization with high resolution. A tree of active heart valves appearance sequences of heart murmurs are obtained by mining locations of active heart valves to model heart murmurs. Due to the heart cyclostationary nature, all sequences of this murmur tree are periodic. Normal heart sound and abnormal heart sound containing various valve regurgitation and stenosis are collected by a rectangular array of 2 × 3 microphones. To evaluate the benefits of the proposed method, it is compared with a murmur classification method based on power spectra density feature extraction. The proposed method achieves 89.8 % accuracy which is superior to mentioned method.

Heart sounds classification using a novel 1-D convolutional neural network with extremely low parameter consumption

Automatic heart sound auscultation is one of the common used techniques for cardiovascular diseases detection. In this paper, a novel heart sound classification method based on deep learning technologies for cardiovascular disease prediction is introduced, which is mainly comprised three parts: pre-processing, 1-D waveform heart sound patches classification using a deep convolutional neural network (CNN) with attention mechanism, and majority voting for final prediction of heart sound recordings. In order to enhance the information flow of the CNNs, a block-stacked style architecture with clique blocks is employed, and in each clique block a bidirectional connection structure is introduced in the proposed CNN. By using the stacked clique and transition blocks, the proposed CNN achieves both spatial and channel attention leading a promising classification performance. Moreover, a novel separable convolution with inverted bottleneck is utilized to decouple the spatial and channel-wise relevancy of features efficiently. Experiments on PhysioNet/CinC 2016 show that the proposed method obtains a superior classification results and excels in consumption of parameter comparing to state-of-the-art methods.

Multichannel lung sound analysis for asthma detection

Background and objectiveLung sound signals convey valuable information of the lung status. Auscultation is an effective technique to appreciate the condition of the respiratory system using lung sound signals. The prior works on asthma detection from lung sound signals rely on the presence of wheeze. In this paper, we have classified normal and asthmatic subjects using advanced signal processing of posterior lung sound signals, even in the absence of wheeze. MethodsWe collected lung sounds of 60 subjects (30 normal and 30 asthma) using a novel 4-channel data acquisition system from four different positions over the posterior chest, as suggested by the pulmonologist. A spectral subband based feature extraction scheme is proposed that works with artificial neural network (ANN) and support vector machine (SVM) classifiers for the multichannel signal. The power spectral density (PSD) is estimated from extracted lung sound cycle using Welch’s method, which then decomposed into uniform subbands. A set of statistical features is computed from each subband and applied to ANN and SVM classifiers to classify normal and asthmatic subjects. ResultsIn the first part of this study, the performances of each individual channel and four channels together are evaluated where the combined channel performance is found superior to that of individual channels. Next, the performances of all possible combinations of the channels are investigated and the best classification accuracies of 89.2( ± 3.87)% and 93.3( ± 3.10)% are achieved for 2-channel and 3-channel combinations in ANN and SVM classifiers, respectively. ConclusionsThe proposed multichannel asthma detection method where the presence of wheeze in lung sound is not a necessary requirement, outperforms commonly used lung sound classification methods in this field and provides significant relative improvement. The channel combination study gives insight into the contribution of respective lung sound collection areas and their combinations in asthma detection.

Detection of pathological heart sounds

Heart sound analysis has been a major topic of research over the past few decades. However, the necessity for a large and reliable database has been a major concern in these studies. Objective: Noting that the current heart sound classification methods do not work properly for noisy signals, the PhysioNet/CinC Challenge 2016 aims to develop the heart sound classification algorithms by providing a global open database for challengers. This paper addresses the problem of heart sound classification methods within noisy real-world phonocardiogram recordings by implementing an innovative approach. Significance: After locating the fundamental heart sounds and the systolic and diastolic components, a novel method named cycle quality assessment is applied to each recording. The presented method detects those cycles which are less affected by noise and better segmented by the use of two criteria here proposed in this paper. The selected cycles are the inputs of a further feature extraction process. Approach: Due to the variability of the heart sound signal induced by various cardiac arrhythmias, four sets of features from the time, time-frequency and perceptual domains are extracted. Before starting the main classification process, the obtained 90-dimensional feature vector is mapped to a new feature space to pre-detect normal recordings by applying a Fisher’s discriminant analysis. The main classification procedure is then done based on three feed-forward neural networks and a voting system among classifiers. Main results: The presented method is evaluated using the training and hidden test sets of the PhysioNet/CinC Challenge 2016. Also, the results are compared with the top five ranked submissions. The results indicate that the proposed method is effective in classifying heart sounds as normal versus abnormal recordings.

Comparison of sound velocity estimation and classification methods for ultrasonic testing of cheese

Testing of two methods novel to ultrasonic measurements was carried out on cheese samples to estimate the Time-of-Flight (TOF) parameter. The Short Time Average/Long Time Average (STA/LTA) method and the Autoregressive Akaike Information Criterion Picker (AR-AIC picker) method are used mainly in seismology for earthquake event detection. The STA/LTA method proved to be ineffective with such noise level that is present during ultrasonic measurements, but the AIC picker algorithm yielded reliable results. A new approach for classification was tested on two types of samples, those were matching in composition, but different in treatment and texture. The method used is based on the results of wavelet decomposition, and after retrieving sufficient spectral data, a linear discriminant analysis (DA) resulted in 100% correct classification, which was compared to the DA classification results based on other methods.