神经网络的声场景自动分类方法*

声场景探察和自动分类能帮助人类制定应对特定环境的正确策略,具有重要的研究价值。随着卷积神经网络的发展,出现了许多基于卷积神经网络的声场景分类方法。其中时频卷积神经网络(TS-CNN)采用了时频注意力模块,是目前声场景分类效果最好的网络之一。为了在保持网络复杂度不变的前提下进一步提高网络的声场景分类性能,该文提出了一种基于协同学习的时频卷积神经网络模型(TSCNN-CL)。具体地说,该文首先建立了基于同构结构的辅助分支参与网络的训练。其次,提出了一种基于KL散度的协同损失函数,实现了分支与主干的知识协同,最后,在测试过程中,为了不增加推理计算量,该文提出的模型只使用主干网络预测结果。在ESC-10、ESC-50和UrbanSound8k数据集的综合实验表明,该模型分类效果要优于TS-CNN模型以及当前大部分的主流方法。     

卷积神经网络用于近红外光谱预测土壤含水率

近红外光谱分析技术在土壤含水率预测方面具有独特的优势,是一种便捷且有效的方法。卷积神经网络作为高性能的深度学习模型,能够从复杂光谱数据中自主提取有效特征结构进行学习,与传统的浅层学习模型相比具有更强的模型表达能力。将卷积神经网络用于近红外光谱预测土壤含水率,并提出了有效的卷积神经网络光谱回归建模方法,简化了光谱数据的预处理要求,且具有更高的光谱预测精度。首先对不同含水率下土壤样品的光谱反射率数据进行简单的预处理,通过主成分分析减少光谱数据量,并将处理后的光谱数据变换为二维光谱信息矩阵,以适应卷积神经网络特殊的学习结构。然后基于卷积神经网络算法,设置双层卷积和池化结构逐层提取光谱数据的内部特征信息,并采用局部连接和权值共享减少网络参数、提高泛化性能。通过试验优化网络结构和各项参数,最终获得针对土壤光谱数据的卷积神经网络土壤含水率预测模型,并与传统的BP,PLSR和LSSVM模型进行对比实验。结果表明在训练样本达到一定数量时,卷积神经网络的预测精度和回归拟合度均高于三种传统模型。在少量训练样本参与建模的情况下,模型预测表现高于BP神经网络,但略低于PLSR和LSSVM模型。随着参与训练样本量的增加,卷积神经网络的预测精度和回归拟合度也随之稳定提升,达到并显著优于传统模型水平。因此,卷积神经网络能够利用近红外光谱数据对土壤含水率做出有效预测,且在较多样本参与建模时取得更好效果。     

A Two-Branch CNN Fusing Temporal and Frequency Features for Motor Imagery EEG Decoding

With the development of technology and the rise of the meta-universe concept, the brain-computer interface (BCI) has become a hotspot in the research field, and the BCI based on motor imagery (MI) EEG has been widely concerned. However, in the process of MI-EEG decoding, the performance of the decoding model needs to be improved. At present, most MI-EEG decoding methods based on deep learning cannot make full use of the temporal and frequency features of EEG data, which leads to a low accuracy of MI-EEG decoding. To address this issue, this paper proposes a two-branch convolutional neural network (TBTF-CNN) that can simultaneously learn the temporal and frequency features of EEG data. The structure of EEG data is reconstructed to simplify the spatio-temporal convolution process of CNN, and continuous wavelet transform is used to express the time-frequency features of EEG data. TBTF-CNN fuses the features learned from the two branches and then inputs them into the classifier to decode the MI-EEG. The experimental results on the BCI competition IV 2b dataset show that the proposed model achieves an average classification accuracy of 81.3% and a kappa value of 0.63. Compared with other methods, TBTF-CNN achieves a better performance in MI-EEG decoding. The proposed method can make full use of the temporal and frequency features of EEG data and can improve the decoding accuracy of MI-EEG.     

Cooperative Spectrum Sensing Based on Multi-Features Combination Network in Cognitive Radio Network

Cognitive radio, as a key technology to improve the utilization of radio spectrum, acquired much attention. Moreover, spectrum sensing has an irreplaceable position in the field of cognitive radio and was widely studied. The convolutional neural networks (CNNs) and the gate recurrent unit (GRU) are complementary in their modelling capabilities. In this paper, we introduce a CNN-GRU network to obtain the local information for single-node spectrum sensing, in which CNN is used to extract spatial feature and GRU is used to extract the temporal feature. Then, the combination network receives the features extracted by the CNN-GRU network to achieve multifeatures combination and obtains the final cooperation result. The cooperative spectrum sensing scheme based on Multifeatures Combination Network enhances the sensing reliability by fusing the local information from different sensing nodes. To accommodate the detection of multiple types of signals, we generated 8 kinds of modulation types to train the model. Theoretical analysis and simulation results show that the cooperative spectrum sensing algorithm proposed in this paper improved detection performance with no prior knowledge about the information of primary user or channel state. Our proposed method achieved competitive performance under the condition of large dynamic signal-to-noise ratio.     

基于改进卷积神经网络算法的语音识别

为了解决传统卷积神经网络识别连续语音数据时识别性能较差的问题,提出一种改进的卷积神经网络算法。该方法引入Fisher准则以及L2正则化约束,在反向传播调整参数阶段,既保证参数误差的最小化,又确保分类以后的样本类间分布较分散,类内分布较集中,同时保证网络权值具有合适的数量级以有效缓解过拟合问题;采用一种更符合生物神经元激活特性的新型log激活函数进行卷积神经网络的优化,进一步提高语音识别的正确率。在语音识别库TIMIT以及THCHS30上的实验结果表明,相较于传统卷积神经网络算法,本文提出的改进算法能较好的提高语音识别率,且泛化能力更强。     

Evaluating the Learning Procedure of CNNs through a Sequence of Prognostic Tests Utilising Information Theoretical Measures

Deep learning has proven to be an important element of modern data processing technology, which has found its application in many areas such as multimodal sensor data processing and understanding, data generation and anomaly detection. While the use of deep learning is booming in many real-world tasks, the internal processes of how it draws results is still uncertain. Understanding the data processing pathways within a deep neural network is important for transparency and better resource utilisation. In this paper, a method utilising information theoretic measures is used to reveal the typical learning patterns of convolutional neural networks, which are commonly used for image processing tasks. For this purpose, training samples, true labels and estimated labels are considered to be random variables. The mutual information and conditional entropy between these variables are then studied using information theoretical measures. This paper shows that more convolutional layers in the network improve its learning and unnecessarily higher numbers of convolutional layers do not improve the learning any further. The number of convolutional layers that need to be added to a neural network to gain the desired learning level can be determined with the help of theoretic information quantities including entropy, inequality and mutual information among the inputs to the network. The kernel size of convolutional layers only affects the learning speed of the network. This study also shows that where the dropout layer is applied to has no significant effects on the learning of networks with a lower dropout rate, and it is better placed immediately after the last convolutional layer with higher dropout rates.     

Learnable three-dimensional Gabor convolutional network with global affinity attention for hyperspectral image classification

Benefiting from the development of hyperspectral imaging technology, hyperspectral image (HSI) classification has become a valuable direction in remote sensing image processing. Recently, researchers have found a connection between convolutional neural networks (CNNs) and Gabor filters. Therefore, some Gabor-based CNN methods have been proposed for HSI classification. However, most Gabor-based CNN methods still manually generate Gabor filters whose parameters are empirically set and remain unchanged during the CNN learning process. Moreover, these methods require patch cubes as network inputs. Such patch cubes may contain interference pixels, which will negatively affect the classification results. To address these problems, in this paper, we propose a learnable three-dimensional (3D) Gabor convolutional network with global affinity attention for HSI classification. More precisely, the learnable 3D Gabor convolution kernel is constructed by the 3D Gabor filter, which can be learned and updated during the training process. Furthermore, spatial and spectral global affinity attention modules are introduced to capture more discriminative features between spatial locations and spectral bands in the patch cube, thus alleviating the interfering pixels problem. Experimental results on three well-known HSI datasets (including two natural crop scenarios and one urban scenario) have demonstrated that the proposed network can achieve powerful classification performance and outperforms widely used machine-learning-based and deep-learning-based methods.     

基于深度神经网络的太阳能电池组件缺陷检测算法研究

针对太阳能电池组件中电池片出现隐裂导致整片电池破碎,最终影响整个组件发电量的问题,在对电池组件光致发光（PL）图像待检测区域筛选定位的基础上,提出了一种利用卷积神经网络（CNN）进行电池组件隐裂缺陷检测的方法。首先利用PL成像方法获取电池组件图像,然后对图像进行预处理,基于聚类的方法对待检测目标区域进行筛选定位,最后利用3种不同结构的卷积神经网络模型对电池片进行缺陷检测,并进行准确率对比,使最优识别准确率达到99.25%。实验结果验证了该方法能准确地检测出太阳能电池组件的隐裂缺陷。     

结合区域提取和改进卷积神经网络的水下小目标检测

针对水下小目标信息量有限而难以提取有效特征导致的检测性能不佳问题,提出了一种结合区域提取和融合Hu矩特征的改进卷积神经网络水下小目标检测方法。该方法包含区域提取和分类两个步骤。首先以马尔可夫随机场分割算法为基础进行区域提取,对潜在目标定位的同时降低伪目标对后续分类的干扰;然后提取潜在目标区域的Hu矩特征并融入卷积神经网络,形成一种形状特征表征能力更强的改进卷积神经网络用于分类。声呐实测数据处理结果表明,该方法可以有效提升对水下小目标的发现概率和正确报警率,与其他目标检测方法相比,该方法具有更好的检测性能和泛化性。     

一种基于2D-CNN的激光超声表面缺陷检测方法

激光超声表面缺陷检测的过程中,缺陷的定量表征通常依赖于操作者的判断,易受到人为因素干扰,致使检测结果不稳定。针对这一问题,提出一种基于图像识别的二维卷积神经网络（2D-CNN）的缺陷自动分类检测方法。利用有限元方法模拟激光超声检测过程,并采集超声信号数据用于训练分类模型;使用连续小变换（CWT）处理超声信号得到小波时频图,以小波时频图作为输入训练卷积神经网络（CNN）分类模型,实现对表面缺陷深度的自动分类。验证结果表明:提出的检测方法能够对不同深度的缺陷准确分类,测试的平均准确率达到97.3%;构建的CNN分类模型能够自主学习输入图像的缺陷特征并完成分类,提高了检测结果稳定性,为激光超声缺陷检测的自动化分析处理提供了新的思路。     

Deep learning and expert knowledge based underwater acoustic OFDM receiver

In this paper, a deep learning and expert knowledge based receiver is proposed for underwater acoustic (UWA) orthogonal frequency division multiplexing (OFDM). Different from the existing deep learning based UWA OFDM receivers, the proposed receiver combines deep learning with the classical expert knowledge of block-based signal processing in UWA OFDM to improve system performance and interpretability. It performs joint channel estimation and signal detection by designing skip connection (SC) convolutional neural network (CNN) cascaded attention mechanism (AM) enhanced bi-directional long short-term memory (BiLSTM) network, abbreviated as SC-CNN-AM-BiLSTM network (SCABNet). Specifically, the channel estimation subnet is designed with SC-CNN to utilize the thought of image super-resolution to reconstruct the entire channel frequency response of all subcarriers. The signal detection subnet is designed with AM-BiLSTM to extract the correlations of received sequential data for signal detection. Especially with the AM, the signal detection subnet can focus more on effective information of the received distorted signal to train the optimal network weights to improve the accuracy of data recovery. The proposed SCABNet is evaluated by experimental data, and the results have demonstrated that the SCABNet has the lowest BER and robust performance compared to the traditional linear algorithm, deep learning based black-box receiver, and ComNet receiver. And the proposed SCABNet is effective and robust when multiple nonideal factors co-exist.     

A Signal processing method of OFDM communication receiver based on CNN

Orthogonal frequency division multiplexing (OFDM) the signal processing is a key issue in wireless communication research. The multipath effect and Doppler shift of wireless communication channels can lead to distortion of the transmitted signal, which poses a considerable challenge to the information recovery of communication receivers. This paper presents the signal processing method of OFDM communication based on convolutional neural network (CNN). The method replaces all signal processing modules of the OFDM communication receiver with CNN, and the information is recovered by the CNN. In order to adapt to the processing of communication signals, we designed a one-dimensional convolutional neural network (1D-CONV-CNN) model as the neural network structures by this method. Simulation results indicate that the signal processing method effectively reduces the bit error rate (BER) and improves its performance compared with the conventional reception method under different channel conditions.     

基于混合神经网络和注意力机制的混沌时间序列预测

为提高混沌时间序列的预测精度,提出一种基于混合神经网络和注意力机制的预测模型(Att-CNNLSTM),首先对混沌时间序列进行相空间重构和数据归一化,然后利用卷积神经网络(CNN)对时间序列的重构相空间进行空间特征提取,再将CNN提取的特征和原时间序列组合,用长短期记忆网络(LSTM)根据空间特征提取时间特征,最后通过注意力机制捕获时间序列的关键时空特征,给出最终预测结果.将该模型对Logistic,Lorenz和太阳黑子混沌时间序列进行预测实验,并与未引入注意力机制的CNN-LSTM模型、单一的CNN和LSTM网络模型、以及传统的机器学习算法最小二乘支持向量机(LSSVM)的预测性能进行比较.实验结果显示本文提出的预测模型预测误差低于其他模型,预测精度更高.     

基于可见/近红外光谱和深度学习的早期鸭胚雌雄信息无损检测

胚蛋雌雄识别一直是家禽业发展的瓶颈问题,在禽肉生产过程中倾向于养殖雄性个体,而禽蛋生产产业倾向于养殖雌性家禽。若能在孵化过程中较早鉴别出种蛋的雌雄,不仅能够降低家禽孵化产业的成本,还能够提高禽蛋和禽肉生产行业的经济效益。该文以种鸭蛋为研究对象,为了在种鸭蛋孵化早期实现对种蛋的雌雄识别,构建了可见/近红外透射光谱信息采集系统,在200～1 100 nm的波长范围内采集了345枚孵化了0~8 d的种鸭蛋光谱数据。搭建了适用于种鸭蛋光谱信息的6层卷积神经网络（convolutional neural network, CNN）,其中包括输入层、3个卷积层、全连接层与输出分类层。卷积层可以提取光谱中的有效信息,全连接层通过对卷积层提取的局部特征进行整合供输出层分类决策。另外在卷积神经网络中引入局部响应归一化和dropout操作能够加快网络的收敛速度。利用该卷积神经网络构建鸭胚雌雄信息识别网络,通过对比与分析不同孵化天数的识别效果,发现孵化7d的识别效果最佳。随后将孵化7 d的种鸭蛋原始光谱数据进行噪声去除,选取500~900 nm波段用于后续的特征波长选取和建模。分别运用了竞争性自适应重加权算法(CARS)、连续投影算法( SPA)与遗传算法(GA)选择能够区分鸭胚性别的波长点,将选取的特征波长转换为二维的光谱信息矩阵,二维光谱信息矩阵保留了一维光谱的有效信息,同时极大地方便了与卷积神经网络的结合。利用二维光谱信息矩阵和卷积神经网络相结合,实现孵化早期阶段鸭胚的雌雄识别。经检验,基于 SPA算法和CNN网络建立的模型效果较佳,其中训练集、开发集及测试集的准确率分别为93.36%,93.12%和93.83%;基于GA算法和CNN网络建立的模型效果次之,训练集、开发集及测试集的准确率分别为90.87%,93.12%和86.42%;基于CARS算法和CNN网络建立的模型的训练集、开发集及测试集的准确率分别为84.65%,83.75%和77.78%。研究结果表明基于可见/近红外光谱技术和卷积神经网络可以实现孵化早期鸭胚胎雌雄的无损鉴别,为后续相关自动化检测装置的研发提供了技术支撑。     

基于CNN的红外图像边缘检测算法的FPGA实现

提出了一种以现场可编程门阵列为硬件处理器实现基于细胞神经网络的红外图像边缘检测方法.首先利用simulink的算法行为特性搭建红外图像输入模块,获得相关的红外图像头信息并对红外图像像素值范围进行相应变化,然后根据细胞神经网络模板所创建的查找表设计单个细胞元软核,再利用细胞神经网络阵列的规则性和互联的局域性,将单个细胞元软核扩展成细胞神经网络阵列.最后采用modelsim将细胞神经网络阵列与红外图像输入、输出模块相关联,从而达到实时处理的效果.实验结果表明:基于现场可编程门阵列为硬件处理器平台实现的细胞神经网络对红外图像进行边缘检测取得了较好的效果,且与MATLAB软件仿真的结果进行对比得出两者只有极其微小的差别.在Xilinx公司Virtex-6系列的现场可编程门阵列平台上,综合后占用极少资源的情况下得到142.693 MHz的最高频率,并且达到了2.378 Mpixels/sec处理速度.     

Binary neutron stars gravitational wave detection based on wavelet packet analysis and convolutional neural networks

This work investigates the detection of binary neutron stars gravitational wave based on convolutional neural network(CNN).To promote the detection performance and efficiency,we proposed a scheme based on wavelet packet(WP)decomposition and CNN.The WP decomposition is a time-frequency method and can enhance the discriminant features between gravitational wave signal and noise before detection.The CNN conducts the gravitational wave detection by learning a function mapping relation from the data under being processed to the space of detection results.This function-mapping-relation style detection scheme can detection efficiency significantly.In this work,instrument effects are con-sidered,and the noise are computed from a power spectral density(PSD)equivalent to the Advanced LIGO design sensitivity.The quantitative evaluations and comparisons with the state-of-art method matched filtering show the excellent performances for BNS gravitational wave detection.On efficiency,the current experiments show that this WP-CNN-based scheme is more than 960 times faster than the matched filtering.     

改进Fast-RCNN的双目视觉车辆检测方法

针对不同空间尺度的车辆表现出显著不同的特征导致检测算法效率低、准确性差且单目难以准确获取车辆距离信息的问题,提出了一种改进Fast-RCNN的汽车目标检测法,利用双目视觉对车辆进行测距。首先利用双目立体相机采集前方图像并进行预处理,加载深度神经网络Fast-RCNN的训练数据,再针对汽车不同空间尺度引入多个内置的子网络,将来自所有子网络的输出自适应组合对车辆进行检测,然后利用SURF特征匹配算法进行左右图像的立体匹配,根据匹配数据进行三维重建并确定车辆质心坐标,从而测量出车辆与双目相机之间的距离。实验结果表明,所述算法可以实现对车辆的快速检测,检测时间比传统的Fast-RCNN缩短了42 ms,并且实现了对5 m范围车辆距离的准确测量,其误差仅为2.4%,精确度高,实时性好。     

基于多输出的 3D 卷积神经网络诊断阿尔兹海默病

随着人口老龄化的加深，阿尔兹海默疾病更加大众化地出现在我们生活中，而早期精准诊断阿尔兹海默疾病并进行正向干预可有效延缓阿尔兹海默疾病的进程．基于磁共振图像的阿尔兹海默疾病的精准诊断需要综合利用多个感兴趣区域（ROIs）的信息，而单个ROI无法体现不同ROIs之间存在的联系与影响．本文首先提出三输入3D卷积神经网络（CNN），综合利用大脑3D磁共振图像中海马体、灰质（无海马体）和白质3个ROIs的信息．此外，随着神经网络的加深，原始图像的重要特征信息会部分丢失，因此我们又提出一种多输出3D CNN，通过增加中间层的连接和输出，缩短输入和输出之间的距离，增强特征传播，减少特征信息的丢失．结果显示采用多输出3DCNN模型实现整个测试集三分类的准确率为90.5%、精确率为91.0%、灵敏度为90.4%、特异性为95.2%、F1-score为90.5%，诊断性能优于单输出3D CNN模型．     

Some optimizations on detecting gravitational wave using convolutional neural network

This work investigates the problem of detecting gravitational wave (GW) events based on simulated damped sinusoid signals contaminated with white Gaussian noise. It is treated as a classification problem with one class for the interesting events. The proposed scheme consists of the following two successive steps: decomposing the data using a wavelet packet, representing the GW signal and noise using the derived decomposition coefficients; and determining the existence of any GW event using a convolutional neural network (CNN) with a logistic regression output layer. The characteristic of this work is its comprehensive investigations on CNN structure, detection window width, data resolution, wavelet packet decomposition and detection window overlap scheme. Extensive simulation experiments show excellent performances for reliable detection of signals with a range of GW model parameters and signal-to-noise ratios. While we use a simple waveform model in this study, we expect the method to be particularly valuable when the potential GW shapes are too complex to be characterized with a template bank.     

Sensor and Actuator Fault Diagnosis for Robot Joint Based on Deep CNN

This paper proposes a data-driven method-based fault diagnosis method using the deep convolutional neural network (DCNN). The DCNN is used to deal with sensor and actuator faults of robot joints, such as gain error, offset error, and malfunction for both sensors and actuators, and different fault types are diagnosed using the trained neural network. In order to achieve the above goal, the fused data of sensors and actuators are used, where both types of fault are described in one formulation. Then, the deep convolutional neural network is applied to learn characteristic features from the merged data to try to find discriminative information for each kind of fault. After that, the fully connected layer does prediction work based on learned features. In order to verify the effectiveness of the proposed deep convolutional neural network model, different fault diagnosis methods including support vector machine (SVM), artificial neural network (ANN), conventional neural network (CNN) using the LeNet-5 method, and long-term memory network (LTMN) are investigated and compared with DCNN method. The results show that the DCNN fault diagnosis method can realize high fault recognition accuracy while needing less model training time.