Research on dynamic routing algorithm based on gaussian mixture model

In this paper, based on the method of environmental sound detection, a neural network model based on capsule network and Gaussian mixture model is proposed. The model proposed in this paper mainly aims at the disadvantages of dynamic routing algorithm in the capsule network, and proposes a dynamic routing algorithm based on Gaussian mixture model. The improved dynamic routing algorithm assumes that the characteristics of the data conform to the multi-dimensional Gaussian distribution, so the model can learn the distribution of data features by building distribution functions of different classes. The information entropy is used as the activation value of the salient degree of the feature. Through experiments, the accuracy of the proposed algorithm on Urbansound8K data set is more than 92%, which is 4.8% higher than the original algorithm.

Comparative sound detection abilities of four decapod crustaceans

Sound perception and detection in decapod crustaceans is surprisingly poorly understood, even though there is mounting evidence for sound playing a critical role in many life history strategies. The suspected primary organ of sound perception are the paired statocysts at the base of the first antennal segment. To better understand the comparative sound detection of decapods, auditory evoked potentials were recorded from the statocyst nerve region of four species (Leptograpsus variegate, Plagusia chabrus, Ovalipes catharus, Austrohelice crassa) in response to two different auditory stimuli presentation methods, shaker table (particle acceleration) and underwater speaker (particle acceleration and pressure). The results showed that there was significant variation in the sound detection abilities between all four species. However, exposure to the speaker stimuli increased all four species sound detection abilities, both in terms of frequency bandwidth and sensitivity, compared to shaker table derived sound detection abilities. This indicates that there is another sensory mechanism in play as well as the statocyst system. Overall, the present research provides comparative evidence of sound detection in decapods and indicates underwater sound detection in this animal group was even more complex than previously thought.

Deep Transfer Learning for Machine Diagnosis: From Sound and Music Recognition to Bearing Fault Detection

Today’s deep learning strategies require ever-increasing computational efforts and demand for very large amounts of labelled data. Providing such expensive resources for machine diagnosis is highly challenging. Transfer learning recently emerged as a valuable approach to address these issues. Thus, the knowledge learned by deep architectures in different scenarios can be reused for the purpose of machine diagnosis, minimizing data collecting efforts. Existing research provides evidence that networks pre-trained for image recognition can classify machine vibrations in the time-frequency domain by means of transfer learning. So far, however, there has been little discussion about the potentials included in networks pre-trained for sound recognition, which are inherently suited for time-frequency tasks. This work argues that deep architectures trained for music recognition and sound detection can perform machine diagnosis. The YAMNet convolutional network was designed to serve extremely efficient mobile applications for sound detection, and it was originally trained on millions of data extracted from YouTube clips. That framework is employed to detect bearing faults for the CWRU dataset. It is shown that transferring knowledge from sound and music recognition to bearing fault detection is successful. The maximum accuracy is achieved using a few hundred data for fine-tuning the fault diagnosis model.

Implementing LED Strips on Road Dividers

Every country suffers from a traffic problem. The reason behind this is not only overpopulation but also heavy vehicles on the road too. The Motive of this paper is to give a better idea to control the traffic and reduce the cost for this working. Using sensors, LED strips, and Rechargeable batteries is enough to implement this system. Today only by hearing the sound of the Ambulance the drivers on the road can leave a way for the Ambulance. It is one of the main reasons why the Ambulance Couldn’t reach on time. The second thing is that heavy vehicles occupy more space on the road. Each road will have at least twolane on both sides. By placing these LED strips in the middle of the road, it could be helpful for the other drivers to leave the way for the Ambulance to come. By implementing this system, we can save time, money, and the valuable life of human beings. Keywords: Sound detection sensor, LED strips, Rechargeable batteries.

Unsupervised Anomalous Sound Detection for Machine Condition Monitoring Using Classification-Based Methods

The task of unsupervised anomalous sound detection (ASD) is challenging for detecting anomalous sounds from a large audio database without any annotated anomalous training data. Many unsupervised methods were proposed, but previous works have confirmed that the classification-based models far exceeds the unsupervised models in ASD. In this paper, we adopt two classification-based anomaly detection models: (1) Outlier classifier is to distinguish anomalous sounds or outliers from the normal; (2) ID classifier identifies anomalies using both the confidence of classification and the similarity of hidden embeddings. We conduct experiments in task 2 of DCASE 2020 challenge, and our ensemble method achieves an averaged area under the curve (AUC) of 95.82% and averaged partial AUC (pAUC) of 92.32%, which outperforms the state-of-the-art models.

SELF-SUPERVISED ACOUSTIC ANOMALY DETECTION VIA CONTRASTIVE LEARNING

We propose an acoustic anomaly detection algorithm based on the framework of contrastive learning. Contrastive learning is a recently proposed self-supervised approach that has shown promising results in image classification and speech recognition. However, its application in anomaly detection is underexplored. Earlier studies have demonstrated that it can achieve state-of-the-art performance in image anomaly detection, but its capability in anomalous sound detection is yet to be investigated. For the first time, we propose a contrastive learning-based framework that is suitable for acoustic anomaly detection. Since most existing contrastive learning approaches are targeted toward images, the effect of other data transformations on the performance of the algorithm is unknown. Our framework learns a representation from unlabeled data by applying audio-specific data augmentations. We show that in the resulting latent space, normal and abnormal points are distinguishable. Experiments conducted on the MIMII dataset confirm that our approach can outperform competing methods in detecting anomalies.

SELF-SUPERVISED ACOUSTIC ANOMALY DETECTION VIA CONTRASTIVE LEARNING

We propose an acoustic anomaly detection algorithm based on the framework of contrastive learning. Contrastive learning is a recently proposed self-supervised approach that has shown promising results in image classification and speech recognition. However, its application in anomaly detection is underexplored. Earlier studies have demonstrated that it can achieve state-of-the-art performance in image anomaly detection, but its capability in anomalous sound detection is yet to be investigated. For the first time, we propose a contrastive learning-based framework that is suitable for acoustic anomaly detection. Since most existing contrastive learning approaches are targeted toward images, the effect of other data transformations on the performance of the algorithm is unknown. Our framework learns a representation from unlabeled data by applying audio-specific data augmentations. We show that in the resulting latent space, normal and abnormal points are distinguishable. Experiments conducted on the MIMII dataset confirm that our approach can outperform competing methods in detecting anomalies.