利用冲击声辨识声源材料的特征提取

声源辨识属于环境声识别的范畴,是模式识别的一个重要研究方向。冲击声携带了大量的声源物理信息,因此利用冲击声提取特征进行声源材料辨识是提高声目标识别分类性能的重要途径。对球-板撞击物理模型合成的冲击声连续统,提出使用基函数学习法提取目标特征,同时利用短时傅里叶变换和小波包变换进行特征提取,以此为基础完成被击平板的材料识别。研究结果表明,利用基函数学习法获得的特征,对于冲击声分类的效果明显优于短时傅里叶变换和小波包变换方法,说明利用该方法进行冲击声声源材料辨识的可行性和优势。      

基于平板冲击声的声源特性表征及自动识别*

以平板结构导纳函数为纽带,建立冲击声信号特征与声源特性之间的关联,获得与声源属性密切相关的特征用于目标分类。针对四边简支矩形被击板,借助信号参数识别算法获得与声源物理属性有关的6维导纳特征,并从冲击声样本中提取80维音色特征,将音色特征和导纳特征做相关性分析,获得与声源物理属性相关的信号特征集。利用BP神经网络进行分类,结果表明,当采用与特定声源物理属性相关的信号特征子集时,分类效果达到同组最优。     

加筋板材料辨识的导纳特征表达与冲击声特征提取

以表征物理属性的导纳特征为中间量,提取与加筋板材料属性有关的冲击声特征。先用相关分析方法获得金属加筋板物理属性的导纳特征表达以及导纳特征与冲击声特征之间的联系,间接得到表征声源物理属性的冲击声特征,然后通过支持向量机分类器验证不同特征在金属加筋板材料分类辨识中的性能。结果表明,所得的4组冲击声特征能准确识别出不同的材料,单个特征的识别率与对应材料属性的可分程度有关,理想冲击声声特征比音色特征的平均识别率更高。由此可见,利用导纳特征提取与材料属性相关冲击声特征的方法是有效的,且所提的特征能够很好的反映声源材料属性。      

一种利用分布式传声器阵列的声源三维定位方法

为了提高噪声和混响条件下分布式传声器阵列进行声源定位的性能,提出一种利用空间稀疏性和压缩感知原理的声源三维定位方法。该方法首先通过两次离散余弦变换方式提取出声音信号特征,并用该特征来构建稀疏定位模型,以便能够综合利用语音信号的短时和长时特性,同时降低模型维数;然后利用在线字典学习技术动态调整字典,克服稀疏模型与实际信号之间的失配问题,增强稀疏定位模型的鲁棒性;进而提出一种改进的平滑l₀范数稀疏重构算法来进行声源位置解算,以提高低信噪比条件下的重构精度。仿真结果表明该方法不仅可以实现多目标定位,而且具有较强的抗噪声和抗混响能力.      

基于Hilbert变换的水下多源声信号频率的相干探测

为实现水下中低频声信号的探测识别,通过研究水下多声源相干探测信号的特征,理论上给出了相干探测信号频谱混叠情况下的特征表达式,并提出了一种基于Hilbert变换的信号解调处理方法,实现了水下多声源相干探测信号频谱混叠情况下各声源发声频率的解调.该方法将探测信号经过滤波平滑处理之后进行Hilbert变换,得到信号的解析形式,然后对解析信号模值的平方进行二次滤波平滑等处理,分离混叠在一起的频带,将得到的信号进行频谱分析,根据频移值计算得到水下各个声源的发声频率.在光学暗室下搭建激光相干探测系统,对2~6kHz的水下声信号进行实验,实验结果表明,该方法可以有效分离探测信号中混叠在一起的信号频带,并准确提取各水下声信号的发声频率,频率提取重复性不大于2.5Hz.     

水下声目标的梅尔倒谱系数智能分类方法

传统水下声目标识别分类方法具有较强的人机交互特性,无法满足未来水下无人平台智能识别分类水声目标的需求。针对这一问题,提出了一种基于梅尔倒谱系数（MFCC）的水下声目标智能识别分类方法,该方法通过提取水下声目标梅尔倒谱系数特征,采用长短时记忆网络（LSTM）构建了智能识别分类模型。使用实际水声信号对该方法进行了验证,结果表明,基于梅尔倒谱系数的水下声目标智能识别分类方法能够在不依赖人工提取特征的情况下,对目标噪声进行识别分类,具备智能化识别分类能力。     

分布式压缩感知麦克风阵列多声源方位估计*

麦克风阵列已被广泛应用于音/视频会议等人机交互领域中时,多声源应用场景对声源方位估计性能提出了更高的要求。压缩感知(CS)声源定位算法将声源定位问题转化为信号的稀疏重构问题,相比传统的定位算法如相位变换加权（SRP-PHAT）和时延累加定位（DS）能够获得较高的定位性能,但多声源的存在一定程度上降低了稀疏程度,影响了CS重构性能。考虑到传统的CS定位算法并未利用多个连续语音帧之间声源空间向量的共同稀疏性,提出采用分布式压缩感知（DCS）理论以改善多声源的稀疏恢复估计的性能。仿真和实验结果表明,相比于传统定位算法和CS-OMP算法,DCS-SOMP算法在不同信噪比和不同声源强度的环境中,对多声源的方位估计都具有更好的定位性能和定位稳健性。     

倒谱参数稀疏分解下的汉语音谎言检测

为了提高汉语语音的谎言检测准确率,提出了一种对信号倒谱参数进行稀疏分解的方法。首先,采用小波包滤波器组对语音信号进行多频带划分,求得子频带对数能量并进行离散余弦变换以提取小波包频带倒谱系数,结合梅尔频率谱系数得到倒谱参数;其次,依据K-奇异值分解方法分别利用说谎和非说谎两种状态下的语音倒谱参数集训练得到过完备混合字典,在此字典上根据正交匹配追踪算法对参数集进行稀疏编码提取稀疏特征;最终进行多种分类模型下的识别实验·实验结果表明,稀疏分解方法相比传统参数降维方法具有更好的优化性能,本文推荐的稀疏谱特征最佳识别率达到78.34%,优于其他特征参数,显著提高了谎言检测识别准确率。      

浅海中利用单水听器的声源被动测距

针对浅海声波导中远距离脉冲声源被动测距问题,提出了一种利用单水听器接收信号自相关函数进行warping变换的声源被动测距方法。理想水下声波导中,接收信号warping变换输出的傅里叶变换频谱中具有不变性频率特征,即与声源距离无关的各简正波截止频率;信号自相关函数中不同简正波相干成分也存在不变性频率特征;推导了未知声源距离时特征频率提取值与不变性频率特征之间的近似关系式。这些规律可推广到实际浅海声波导,并用于声源被动测距。利用声场计算模型来提供具有不变性频率特征的频谱,对2011年12月北黄海海域水声实验中单水听器接收的脉冲声数据进行了处理,验证了方法的有效性,测距结果和实际距离符合良好,平均测距误差在10%以内。      

深海声影区稀疏时延估计与声源测距

研究了深海声影区中经一次海底反射的多途声线到达垂直双水听器的时延差与声源位置的关系,提出了一种稀疏时延估计与声源测距方法。首先利用近海面布放的短间距垂直双水听器接收一定频带的声信号,然后计算接收信号的广义互相关函数,并利用频谱搬移和稀疏解卷积技术提取时延差,最后通过时延差匹配,估计水下声源的距离。仿真实验表明,在4300 m深海中,所提方法能够正确提取多途到达时延差,估计声影区内的声源距离。海试结果表明,当垂直接收孔径分别为21 m和30 m时,声源测距误差分别小于13.6%和8.1%。上述结果表明,所提出的时延估计方法可适应带宽较窄的接收信号,多途到达时延估计参数可用于实现声影区中的水下声源测距。      

联合稀疏表示在水下目标分类中的应用

水下目标分类识别的性能受所选特征的限制,多特征往往可以获得更加稳定的结果,针对这一问题,提出了一种基于联合稀疏表示模型的水下目标分类识别方法。首先对水下目标回波信号提取3种具有信息互补性与关联性的特征:中心矩特征、小波包能量谱特征、梅尔频率倒谱系数特征,然后应用加速近端梯度法对联合稀疏表示模型进行优化,求解得到最优联合稀疏系数,最后根据最小误差准则确定目标类别。在消声水池开展模拟实验,对6类目标进行分类识别,结果表明:与传统算法相比,提出的算法具有更高识别准确率,并且其执行效率较传统算法有很大提升。      

Research on Channel Selection and Multi-Feature Fusion of EEG Signals for Mental Fatigue Detection

With the rapid development of modern social science and technology, the pace of life is getting faster, and brain fatigue has become a sub-health state that seriously affects the normal life of people. Electroencephalogram (EEG) signals reflect changes in the central nervous system. Using EEG signals to assess mental fatigue is a research hotspot in related fields. Most existing fatigue detection methods are time-consuming or don’t achieve satisfactory results due to insufficient features extracted from EEG signals. In this paper, a 2-back task is designed to induce fatigue. The weight value of each channel under a single feature is calculated by ReliefF algorithm. The classification accuracy of each channel under the corresponding features is analyzed. The classification accuracy of each single channel is combined to perform weighted summation to obtain the weight value of each channel. The first half channels sorted in descending order based on the weight value is chosen as the common channels. Multi-features in frequency and time domains are extracted from the common channel data, and the sparse representation method is used to perform feature fusion to obtain sparse fused features. Finally, the SRDA classifier is used to detect the fatigue state. Experimental results show that the proposed methods in our work effectively reduce the number of channels for computation and also improve the mental fatigue detection accuracy.     

Feature extraction for sound source material recognition with impact sounds

Sound source recognition is a part of environmental sound recognition,which is one of the most important research areas in pattern recognition.Impact sounds carry much physical information associated with the sound sources,which makes impact sound based sound source recognition an important approach to improve recognition performance.In this study,the impact sound continuum synthesized with a ball-plate collision model is used for material recognition of the impacted plates.The basis function learning method and time-frequency representation methods,including the short time Fourier transform and the wavelet packet transform,are applied into classification and the recognition results are compared.The result shows that the features obtained by using the basis function learning perform better for material classification of the impacted plates than that by using the short time Fourier transform and the wavelet packet transform.This demonstrates the high efficiency and superiority of this method in material recognition of sound sources.     

Analysis of acoustic parameters for consonant voicing classification in clean and telephone speech

This paper describes acoustic cues for classification of consonant voicing in a distinctive feature-based speech recognition system. Initial acoustic cues are selected by studying consonant production mechanisms. Spectral representations, band-limited energies, and correlation values, along with Mel-frequency cepstral coefficients features (MFCCs) are also examined. Analysis of variance is performed to assess relative significance of features. Overall, 82.2%, 80.6%, and 78.4% classification rates are obtained on the TIMIT database for stops, fricatives, and affricates, respectively. Combining acoustic parameters with MFCCs shows performance improvement in all cases. Also, performance in the NTIMIT telephone channel speech shows that acoustic parameters are more robust than MFCCs.     

A Non-Contact Fault Diagnosis Method for Bearings and Gears Based on Generalized Matrix Norm Sparse Filtering

Fault diagnosis of mechanical equipment is mainly based on the contact measurement and analysis of vibration signals. In some special working conditions, the non-contact fault diagnosis method represented by the measurement of acoustic signals can make up for the lack of contact testing. However, its engineering application value is greatly restricted due to the low signal-to-noise ratio (SNR) of the acoustic signal. To solve this deficiency, a novel fault diagnosis method based on the generalized matrix norm sparse filtering (GMNSF) is proposed in this paper. Specially, the generalized matrix norm is introduced into the sparse filtering to seek the optimal sparse feature distribution to overcome the defect of low SNR of acoustic signals. Firstly, the collected acoustic signals are randomly overlapped to form the sample fragment data set. Then, three constraints are imposed on the multi-period data set by the GMNSF model to extract the sparse features in the sample. Finally, softmax is used to as a classifier to categorize different fault types. The diagnostic performance of the proposed method is verified by the bearing and planetary gear datasets. Results show that the GMNSF model has good feature extraction ability performance and anti-noise ability than other traditional methods.     

融合自适应稀疏表示和相关系数的高光谱伪装分类方法

目前比较成熟的高光谱成像手段有卫星遥感和航空成像技术,这两种成像方式侦察时间大致相同,入射光方向基本一致,因而地物的光谱曲线比较固定;在陆基条件下,地物的光谱曲线受成像环境的影响凸显,因此应该对适用于陆基条件下的高光谱图像分类方法进行研究。在陆基高光谱图像中,对每个地物进行类型以及种类的判别有利于后续对目标的识别和处理,不同于传统遥感图像分类,陆基条件下的高光谱图像目标分类训练样本不仅较难获得,并且在陆基条件下的高光谱图像中,训练样本之间的相关性随着目标类型、探测器参数以及成像环境等因素时刻发生变化。基于稀疏性表示的分类方法已经被广泛应用于处理图像问题以及各种机器视觉问题。对于陆基高光谱图像来说,基于固定范数约束的稀疏编码策略无法适应陆基条件下高光谱成像多变的环境,而自适应稀疏表示可以根据样本相关性自适应的调节范数约束,相关系数可以提高图像中的破坏因素（阴影、噪声点等）的识别精度。通过引入正则化参数,融合了自适应稀疏表示和相关系数,提出了一种新的高光谱图像分类方法。为了验证所提方法的有效性,分别在绿色植被背景和荒漠背景中设置伪装物,通过不同的分类方法对图像进行分类,实验结果表明,不管是分类精度还是分类一致性,该方法都有明显的优势,可以应用于陆基条件下的高光谱图像分类,为目标分类提供了理论基础。     

结合卷积神经网络的浅海有源探测信道匹配

信道匹配方法在有源探测领域是一种重要的提升检测信噪比的方法。针对非确知海底参数环境下的有源探测信道匹配问题,提出一种结合卷积神经网络进行信道匹配的算法。该算法基于海底参数扰动开展声场仿真生成卷积网络训练数据;首先通过分类网络将信号按照海底底质类型分类,在每个分类区间内采用单独的卷积网络反演海底参数;然后结合声场模型估计信道传递函数,进行信道匹配,从而在非确知环境下抑制多途影响,提升回波检测能力。仿真与实验结果表明,该算法能够在不确知海底环境条件下,有效估计信道传递函数,实现信道最优化匹配,在实验条件下可提高回波检测信噪比4 dB左右。相比传统方法,该算法可以在海底参数不确知条件下对低接收信噪比的信号实现信道匹配,同时不需要高信噪比的实验参考信号,有效提高了信道匹配方法的环境宽容性。      

Evaluation of formant-like features on an automatic vowel classification task

Numerous attempts have been made to find low-dimensional, formant-related representations of speech signals that are suitable for automatic speech recognition. However, it is often not known how these features behave in comparison with true formants. The purpose of this study was to compare two sets of automatically extracted formant-like features, i.e., robust formants and HMM2 features, to hand-labeled formants. The robust formant features were derived by means of the split Levinson algorithm while the HMM2 features correspond to the frequency segmentation of speech signals obtained by two-dimensional hidden Markov models. Mel-frequency cepstral coefficients (MFCCs) were also included in the investigation as an example of state-of-the-art automatic speech recognition features. The feature sets were compared in terms of their performance on a vowel classification task. The speech data and hand-labeled formants that were used in this study are a subset of the American English vowels database presented in Hillenbrand et al. [J. Acoust. Soc. Am. 97, 3099-3111 (1995)]. Classification performance was measured on the original, clean data and in noisy acoustic conditions. When using clean data, the classification performance of the formant-like features compared very well to the performance of the hand-labeled formants in a gender-dependent experiment, but was inferior to the hand-labeled formants in a gender-independent experiment. The results that were obtained in noisy acoustic conditions indicated that the formant-like features used in this study are not inherently noise robust. For clean and noisy data as well as for the gender-dependent and gender-independent experiments the MFCCs achieved the same or superior results as the formant features, but at the price of a much higher feature dimensionality.     

A sparse representation method based on kernel and virtual samples for face recognition

To improve the classification accuracy of face recognition, a sparse representation method based on kernel and virtual samples is proposed in this paper. The proposed method has the following basic idea: first, it extends the training samples by copying the left side of the original training samples to the right side to form virtual training samples. Then the virtual training samples and the original training samples make up a new training set and we use a kernel-induced distance to determine M nearest neighbors of the test sample from the new training set. Second, it expresses the test sample as a linear combination of the selected M nearest training samples and finally exploits the determined linear combination to perform classification of the test sample. A large number of face recognition experiments on different face databases illustrate that the error ratios obtained by our method are always lower more or less than face recognition methods including the method mentioned in Xu and Zhu [21], the method proposed in Xu and Zhu [39], sparse representation method based on virtual samples (SRMVS), collaborative representation based classification with regularized least square (CRC_RLS), two-phase test sample sparse representation (TPTSSR), and the feature space-based representation method.     

Classification of stop place in consonant-vowel contexts using feature extrapolation of acoustic-phonetic features in telephone speech

Knowledge-based speech recognition systems extract acoustic cues from the signal to identify speech characteristics. For channel-deteriorated telephone speech, acoustic cues, especially those for stop consonant place, are expected to be degraded or absent. To investigate the use of knowledge-based methods in degraded environments, feature extrapolation of acoustic-phonetic features based on Gaussian mixture models is examined. This process is applied to a stop place detection module that uses burst release and vowel onset cues for consonant-vowel tokens of English. Results show that classification performance is enhanced in telephone channel-degraded speech, with extrapolated acoustic-phonetic features reaching or exceeding performance using estimated Mel-frequency cepstral coefficients (MFCCs). Results also show acoustic-phonetic features may be combined with MFCCs for best performance, suggesting these features provide information complementary to MFCCs.