SLIC Superpixel-Based l2,1-Norm Robust Principal Component Analysis for Hyperspectral Image Classification

Dimensionality Reduction of Hyperspectral Images Based on Improved Spatial-Spectral Weight Manifold Embedding

Due to the spectral complexity and high dimensionality of hyperspectral images (HSIs), the processing of HSIs is susceptible to the curse of dimensionality. In addition, the classification results of ground truth are not ideal. To overcome the problem of the curse of dimensionality and improve classification accuracy, an improved spatial-spectral weight manifold embedding (ISS-WME) algorithm, which is based on hyperspectral data with their own manifold structure and local neighbors, is proposed in this study. The manifold structure was constructed using the structural weight matrix and the distance weight matrix. The structural weight matrix was composed of within-class and between-class coefficient representation matrices. These matrices were obtained by using the collaborative representation method. Furthermore, the distance weight matrix integrated the spatial and spectral information of HSIs. The ISS-WME algorithm describes the whole structure of the data by the weight matrix constructed by combining the within-class and between-class matrices and the spatial-spectral information of HSIs, and the nearest neighbor samples of the data are retained without changing when embedding to the low-dimensional space. To verify the classification effect of the ISS-WME algorithm, three classical data sets, namely Indian Pines, Pavia University, and Salinas scene, were subjected to experiments for this paper. Six methods of dimensionality reduction (DR) were used for comparison experiments using different classifiers such as k-nearest neighbor (KNN) and support vector machine (SVM). The experimental results show that the ISS-WME algorithm can represent the HSI structure better than other methods, and effectively improves the classification accuracy of HSIs.

Hyperspectral Superpixel-Wise Glioblastoma Tumor Detection in Histological Samples

The combination of hyperspectral imaging (HSI) and digital pathology may yield more accurate diagnosis. In this work, we propose the use of superpixels in HS images for combining regions of pixels that can be classified according to their spectral information to classify glioblastoma (GB) brain tumors in histologic slides. The superpixels are generated by a modified simple linear iterative clustering (SLIC) method to accommodate HS images. This work employs a dataset of H&E (Hematoxylin and Eosin) stained histology slides from 13 patients with GB and over 426,000 superpixels. A linear support vector machine (SVM) classifier was performed on independent training, validation, and testing datasets. The results of this investigation show that the proposed method can detect GB brain tumors from non-tumor samples with average sensitivity and specificity of 87% and 81%, respectively. The overall accuracy of this method is 83%. The study demonstrates that hyperspectral digital pathology can be useful for detecting GB brain tumors by exploiting spectral information alone on a superpixel level.

A Novel Spectral–Spatial Classification Method for Hyperspectral Image at Superpixel Level

Although superpixel segmentation provides a powerful tool for hyperspectral image (HSI) classification, it is still a challenging problem to classify an HSI at superpixel level because of the characteristics of adaptive size and shape of superpixels. Furthermore, these characteristics of superpixels along with the appearance of noisy pixels makes it difficult to appropriately measure the similarity between two superpixels. Under the assumption that pixels within a superpixel belong to the same class with a high probability, this paper proposes a novel spectral–spatial HSI classification method at superpixel level (SSC-SL). Firstly, a simple linear iterative clustering (SLIC) algorithm is improved by introducing a new similarity and a ranking technique. The improved SLIC, specifically designed for HSI, can straightly segment HSI with arbitrary dimensionality into superpixels, without consulting principal component analysis beforehand. In addition, a superpixel-to-superpixel similarity is newly introduced. The defined similarity is independent of the shape of superpixel, and the influence of noisy pixels on the similarity is weakened. Finally, the classification task is accomplished by labeling each unlabeled superpixel according to the nearest labeled superpixel. In the proposed superpixel-level classification scheme, each superpixel is regarded as a sample. This obviously greatly reduces the data volume to be classified. The experimental results on three real hyperspectral datasets demonstrate the superiority of the proposed spectral–spatial classification method over several comparative state-of-the-art classification approaches, in terms of classification accuracy.

Rapid Recognition of Field-Grown Wheat Spikes Based on a Superpixel Segmentation Algorithm Using Digital Images

Wheat spike number, which could be rapidly and accurately estimated by the image processing technology, serves as the basis for crop growth monitoring and yield prediction. In this research, simple linear iterative clustering (SLIC) was performed for superpixel segmentation of the digital images of field-grown wheat. Firstly, certain characteristic color parameters were extracted and analyzed from the digital images, and the classifiers with the highest accuracy were chosen for subsequent image classification. Next, the main body of wheat spike was extracted through a series of morphological transformation and estimate was performed for each region. Backbone of the head was extracted, and the number of inflection points of backbone was detected. Then the wheat spike number was determined by combining the estimate of inflection points of backbone and the estimate for each region. Finally, the wheat spike number estimate was verified under four nitrogen fertilizer levels. The results were as follows: (1) Super green value (Eg) and normalized red green index (Dgr) were used as classification features to recognize wheat spikes, soil and leaves; (2) compared with pixel-based image processing, wheat spike recognition effect was much better after superpixel segmentation, as the main body of wheat spike extracted was more clear and morphology more intact; and (3) wheat plants had better growth under high nitrogen fertilizer level, and the accuracy of wheat spike number estimation was also the highest, which was 94.01%. The growth status was the worst under no nitrogen fertilizer application, and the accuracy of wheat spikes number estimation was also the lowest, which was only 80.8%. After excluding the no nitrogen condition, the accuracy of wheat spikes number estimation among mixed samples with more uniform growth status was up to 93.8%, which was an increase by 10.1% than before the exclusion. Wheat spikes number estimate based on superpixel segmentation and color features was a rapid and accurate method that was applicable to the field environment. However, this method was not recommended for use when the growth status of wheat was poor or of high heterogeneity. The findings provided reference for field-grown wheat yield estimate.