Change Detection for Hyperspectral Images Using Extended Mutual Information and Oversegmentation

Taşkesen, Bahar; Koz, Alper; Alatan, A. Aydın; Weatherbee, Oliver

doi:10.1109/whispers.2018.8747018

Cited by 8 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two phases are applied over the images namely high segmentation and joint super pixel with suitable kernels. 6 The analysis of the infrared images is done by changing the location, altitude and time parameter in radiance signature. It uses different targets and environments for making the model more effective.…”

Section: Related Workmentioning

confidence: 99%

Object detection and estimation: A hybrid image segmentation technique using convolutional neural network model

Sundaram

Sakthivel

2022

Concurrency and Computation

View full text Add to dashboard Cite

Object detection from image is more challenging and integral part in the inter-discipline area of computer vision. The computer vision is highly attractive in many applications like human pose estimation, instance segmentation, recognizing action, disease predictions object prediction and many more applications. The traditional method of detecting objects from the images is done using bounding boxes with labels. It suffers from the overlapping of the boxes with various smaller objects, which leads to accuracy issues in detection problems. Hence, machine learning techniques are used to detect the relevant objects from the image using center point to avoid the nonmaximal suppression in bounding box. To accurately identify images, an U-Net architecture based object detection method is proposed. In this model, it effectively uses semantic level segmentation and instance segmentation. This system effectively identifies all the objects present in the given image using the efficient hybrid segmentation models and Gromov Hausdroff distance measure. For experimentation, two data sets are used for evaluation of the model to identify all categories of objects from the image. The proposed model achieves an accuracy of 91.8% and reliable when compared to existing effective object detection algorithms like fully convolution network (FCN), YOLO (you only look once) and mask region based-convolutional neural network (mask R-CNN) model.

show abstract

Section: Related Workmentioning

confidence: 99%

Object detection and estimation: A hybrid image segmentation technique using convolutional neural network model

Sundaram

Sakthivel

2022

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Here, super-pixels are image regions, built from pixels that share common information and that are relatively close in the spatial dimension. An over-segmentation algorithm produces an overestimated number of super-pixels that might be joined together, because of their closeness, producing real segments [91,92].…”

Section: Hyperspectral Image Clusteringmentioning

confidence: 99%

A Robust Stochastic Approach to Mineral Hyperspectral Analysis for Geometallurgy

et al. 2020

View full text Add to dashboard Cite

Most mining companies have registered important amounts of drill core composite spectra using different acquisition equipment and by following diverse protocols. These companies have used classic spectrography based on the detection of absorption features to perform semi-quantitative mineralogy. This methodology requires ideal laboratory conditions in order to obtain normalized spectra to compare. However, the inherent variability of spectral features—due to environmental conditions and geological context, among others—is unavoidable and needs to be managed. This work presents a novel methodology for geometallurgical sample characterization consisting of a heterogeneous, multi-pixel processing pipeline which addresses the effects of ambient conditions and geological context variability to estimate critical geological and geometallurgical variables. It relies on the assumptions that the acquisition of hyperspectral images is an inherently stochastic process and that ore sample information is deployed in the whole spectrum. The proposed framework is basically composed of: (a) a new hyperspectral image segmentation algorithm, (b) a preserving-information dimensionality reduction scheme and (c) a stochastic hierarchical regression model. A set of experiments considering white reference spectral characterization and geometallurgical variable estimation is presented to show promising results for the proposed approach.

show abstract

“…Change detection technology is widely used in remote sensing applications such as urban planning, natural disaster management, agricultural investigation, and ecosystem monitoring [10]. The complete hyperspectral image change detection process roughly includes the following steps: (1) Hyperspectral image data acquisition; (2) Hyperspectral image preprocessing: after preprocessing, multi-temporal images can ensure their spatial comparability; (3) Generating difference maps and extracting difference information from images in different phases, which directly affects how to design an effective change detection algorithm; (4) Result evaluation: the performance of the proposed change detection algorithm is evaluated by some evaluation criteria [11].…”

Section: Introductionmentioning

confidence: 99%

MMSRC: A Multidirection Multiscale Spectral–Spatial Residual Network for Hyperspectral Multiclass Change Detection

Tang

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Recently, deep convolutional neural network (CNN) hyperspectral change detection methods have achieved significant improvement. However, most CNN hyperspectral change detection methods do not make full use of spectral-spatial feature information. In this paper, we propose a novel multi-direction and multi-scale spectral-spatial residual network for hyperspectral multi-class change detection. Specifically, multi-scale structure and a multi-direction mechanism are introduced to investigate feature variation of hyperspectral images and improve the accuracy of hyperspectral change detection. Experiments on multiple hyperspectral datasets show that the proposed method achieves improved performance in comparison with other advanced hyperspectral multi-class change detection methods.

show abstract

Change Detection for Hyperspectral Images Using Extended Mutual Information and Oversegmentation

Cited by 8 publications

References 11 publications

Object detection and estimation: A hybrid image segmentation technique using convolutional neural network model

Object detection and estimation: A hybrid image segmentation technique using convolutional neural network model

A Robust Stochastic Approach to Mineral Hyperspectral Analysis for Geometallurgy

MMSRC: A Multidirection Multiscale Spectral–Spatial Residual Network for Hyperspectral Multiclass Change Detection

Contact Info

Product

Resources

About