No abstract

Anomaly detection in the sound from machines is an important task in machine monitoring. An autoencoder architecture based on the reconstruction error using a log-Mel spectrogram feature is a conventional approach for this domain. However, because of the non-stationary nature of some sounds from the target machine, such a conventional approach does not perform well in those circumstances. In this paper, we propose a novel approach regarding the choice of used features and a new auto-encoder architecture. We created the Mixed Feature, which is a mixture of different sound representations, and a new deep learning method called Fully-Connected U-Net, a form of autoencoder architecture. With experiments on the same dataset as the baseline system, using the same architecture for all types of machines, the experimental results showed that our methods outperformed the baseline system in terms of the AUC and pAUC evaluation metrics. The optimized model achieved 83.38% AUC and 64.51% pAUC on average overall machine types on the developed dataset and outperformed the published baseline by 13.43% AUC and 8.13% pAUC.

Recently years, vector map has many advantages than raster map when it is used in many domains as research, education, military or digital map services. And in most cases vector map data contains confidential information which must be kept away from unauthorized users. Moreover, the producing process of a vector map is considerably complex, and the maintenance of a digital map requires substantial monetary, human resources. With the rapid development of vector map contents, a large volume of valuable vector map dataset has been illegal distributed by pirates, hackers or unauthorized users. Therefore the problem focuses on how to protect the vector map data for multimedia applications, storage and transmission. This paper presents the selective encryption algorithm for vector map protection for storage, transmission, distribution to authorized users. In proposed algorithm, we just select some values of polylines and polygons in DCT domain to encrypt by random algorithms and cryptography. Experimental results verified that proposed algorithm is effectively and security. Maps are changed whole after encryption process, and unauthorized users cannot access to copy or use them. Encrypted maps do not alter the size of file and it does not have loss accuracy. The error between original map and decrypted map is approximate zero

No abstract

No abstract