Hyperspectral anomaly detection (HAD) aims to discern the objects deviated dramatically from their surrounding pixels. Some deep learning-based models integrating with the low-rank representation (LRR) have been propo...
详细信息
Anomaly detection refers to the identification of cases that do not conform to the expected pattern, which takes a key role in diverse research areas and application domains. Most of existing methods can be summarized...
详细信息
Anomaly detection refers to the identification of cases that do not conform to the expected pattern, which takes a key role in diverse research areas and application domains. Most of existing methods can be summarized as anomaly object detection-based and reconstruction error-based techniques. However, due to the bottleneck of defining encompasses of real-world high-diversity outliers and inaccessible inference process, individually, most of them have not derived groundbreaking progress. To deal with those imperfectness, and motivated by memory-based decision-making and visual attention mechanism as a filter to select environmental information in human vision perceptual system, in this paper, we propose a Multi-scale Attention memory with hash addressing autoencoder network (MAMA Net) for anomaly detection. First, to overcome a battery of problems result from the restricted stationary receptive field of convolution operator, we coin the multi-scale global spatial attention block which can be straightforwardly plugged into any networks as sampling, upsampling and downsampling function. On account of its efficient features representation ability, networks can achieve competitive results with only several level blocks. Second, it's observed that traditional autoencoder can only learn an ambiguous model that also reconstructs anomalies "well" due to lack of constraints in training and inference process. To mitigate this challenge, we design a hash addressing memory module that proves abnormalities to produce higher reconstruction error for classification. In addition, we couple the mean square error (MSE) with Wasserstein loss to improve the encoding data distribution. Experiments on various datasets, including two different COVID-19 datasets and one brain MRI (RIDER) dataset prove the robustness and excellent generalization of the proposed MAMA Net.
Diesel engines often present challenges in achieving efficient warning and accurate localisation during actual operation due to complex working conditions and the lack of fault samples. To address this issue, this pap...
详细信息
Diesel engines often present challenges in achieving efficient warning and accurate localisation during actual operation due to complex working conditions and the lack of fault samples. To address this issue, this paper proposes a zero-fault sample anomaly warning and localisation method for diesel engines, which is based on multi-cylinder similarity and memory augmented autoencoder. This approach allows for model training without the use of fault samples, a method that has been scarcely explored in industrial research concerning diesel engine fault diagnosis using only normal samples. The method leverages the similarities in operating states among multiple cylinders and constructs a joint similarity matrix by integrating waveform similarity and impact energy similarity. This approach captures the intrinsic correlation characteristics among the cylinders and effectively mitigates the influence of changes in operating conditions on diagnostic performance. Additionally, a memory augmentation autoencoder is employed to model normal samples. By amplifying the reconstruction error of abnormal data, the method facilitates fault warning and localisation even in the absence of fault samples. Experimental results demonstrate that the proposed method can successfully achieve early fault warning and localisation under both single and cross-operating conditions without requiring fault samples, thereby providing an effective solution for early fault detection and intelligent maintenance of diesel engines.
Recently,the autoencoder(AE)based method plays a critical role in the hyperspectral anomaly detection ***,due to the strong generalised capacity of AE,the abnormal samples are usually reconstructed well along with the...
详细信息
Recently,the autoencoder(AE)based method plays a critical role in the hyperspectral anomaly detection ***,due to the strong generalised capacity of AE,the abnormal samples are usually reconstructed well along with the normal background ***,in order to separate anomalies from the background by calculating reconstruction errors,it can be greatly beneficial to reduce the AE capability for abnormal sample reconstruction while maintaining the background reconstruction performance.A memory‐augmented autoencoder for hyperspectral anomaly detection(MAENet)is proposed to address this challenging ***,the proposed MAENet mainly consists of an encoder,a memory module,and a ***,the encoder transforms the original hyperspectral data into the low‐dimensional latent ***,the latent representation is utilised to retrieve the most relevant matrix items in the memory matrix,and the retrieved matrix items will be used to replace the latent representation from the ***,the decoder is used to reconstruct the input hyperspectral data using the retrieved memory *** this strategy,the background can still be reconstructed well while the abnormal samples *** conducted on five real hyperspectral anomaly data sets demonstrate the superiority of the proposed method.
暂无评论