Multiscale anomaly detection using diffusion maps and saliency score

Mishne, Gal; Cohen, Israel

doi:10.1109/icassp.2014.6854115

Cited by 11 publications

(38 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The diffusion distance between (x, y) is small if there is a large number of short paths connecting them in the graph. The diffusion distance is robust to noise because the distance between (x, y) depends on all possible paths between the points within the dataset [9,10]. Combining Equations (10) and (11), D t (x, y) can be defined by…”

Section: Diffusion Mapmentioning

confidence: 99%

“…Because the pixels in the background are clustered together, the shadowed and target areas would be distant from this cluster in the new embedding. Using the target scoring equation defined from the diffusion distance [10], the target and shadowed areas of an SSS image can be detected.…”

Section: Diffusion Mapmentioning

confidence: 99%

“…When the DM is constructed, and the diffusion distance provides metrics for target detection, the dissimilarity measure between two image patches can be defiened by [10] …”

Section: The Ns + Dm Algorithm For Sss Image Target Detectionmentioning

confidence: 99%

“…DMs have also been used to solve computer vision problems such as mine-like object detection and recognition for SSS images [9,10]. A DM is a powerful tool for handling large datasets, which is helpful for processing SSS images with high resolution.…”

Section: Introductionmentioning

confidence: 99%

“…Grasso et al [8] proposed a small target detection method based on local gray level information and a mathematical morphology operation, but the method only fitted those SSS images obtained by an automated underwater vehicle (AUV) platform. Mishne et al [9,10] studied an image anomaly region detection method based on a diffusion map, but the method did not account for the problems caused by the shadowed areas of a target, the complexity of the noise, or the features of the sea bottom being influenced in the SSS image. Fakiris et al [11] and Rhinelander et al [12] proposed an automatic target detection and recognition method based on SSS image texture features, which is also a supervised method that requires sample data, but robust texture features and sufficient samples are difficult to obtain, especially for SSS images obtained in complex marine noise.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Side Scan Sonar Image Target Detection Algorithm Based on a Neutrosophic Set and Diffusion Maps

et al. 2018

View full text Add to dashboard Cite

Abstract:To accurately achieve side scan sonar (SSS) image target detection, a novel target detection algorithm based on a neutrosophic set (NS) and diffusion maps (DMs) is proposed in this paper. Firstly, the neutrosophic subset images were obtained by transforming the input SSS image into the NS domain. Secondly, the shadowed areas of the SSS image were detected using the single gray value threshold method before the diffusion map was calculated. Lastly, based on the diffusion map, the target areas were detected using the improved target scoring equation defined by the diffusion distance and texture feature. The experiments using SSS images of single clear and unclear targets, with or without shadowed areas, showed that the algorithm accurately detects targets. Experiments using SSS images of multiple targets, with or without shadowed areas, showed that no false or missing detections occurred. The target areas were also accurately detected in SSS images with complex features such as sand wave terrain. The accuracy and effectiveness of the proposed algorithm were assessed.

show abstract

Section: Diffusion Mapmentioning

confidence: 99%

Section: Diffusion Mapmentioning

confidence: 99%

“…When the DM is constructed, and the diffusion distance provides metrics for target detection, the dissimilarity measure between two image patches can be defiened by [10] …”

Section: The Ns + Dm Algorithm For Sss Image Target Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Side Scan Sonar Image Target Detection Algorithm Based on a Neutrosophic Set and Diffusion Maps

et al. 2018

View full text Add to dashboard Cite

show abstract

Image Anomalies: A Review and Synthesis of Detection Methods

et al. 2019

View full text Add to dashboard Cite

We review the broad variety of methods that have been proposed for anomaly detection in images. Most methods found in the literature have in mind a particular application. Yet we focus on a classification of the methods based on the structural assumption they make on the "normal" image, assumed to obey a "background model". Five different structural assumptions emerge for the background model. Our analysis leads us to reformulate the best representative algorithms in each class by attaching to them an a-contrario detection that controls the number of false positives and thus deriving a uniform detection scheme for all. By combining the most general structural assumptions expressing the background's normality with the proposed generic statistical detection tool, we end up proposing several generic algorithms that seem to generalize or reconcile most methods. We compare the six best representatives of our proposed classes of algorithms on anomalous images taken from classic papers on the subject, and on a synthetic database. Our conclusion hints that it is Both authors contributed equally to this work. possible to perform automatic anomaly detection on a single image.

show abstract

Supervised anomaly detection by convolutional sparse representation

Pourhashemi

Mahmoudzadeh

2022

Multimed Tools Appl

View full text Add to dashboard Cite

Multiscale anomaly detection using diffusion maps and saliency score

Cited by 11 publications

References 23 publications

A Side Scan Sonar Image Target Detection Algorithm Based on a Neutrosophic Set and Diffusion Maps

A Side Scan Sonar Image Target Detection Algorithm Based on a Neutrosophic Set and Diffusion Maps

Image Anomalies: A Review and Synthesis of Detection Methods

Supervised anomaly detection by convolutional sparse representation

Contact Info

Product

Resources

About