Combining spectral and texture features in hyperspectral image analysis for plant monitoring

Abstract

A texture enhanced spectral analysis framework is proposed for classifying hyperspectral images of plants of different conditions. Differentiating different plant conditions is important to precision agriculture as it helps detect diseases and stresses and optimise growth control. Advanced machine learning techniques are used to identify distinctive features in the spectral domain of hyperspectral images. In addition, texture properties are explored in the subband images. The framework integrates these two levels of properties at both feature extraction and classifying decision stages. The main crux of the work lies in the use of the significant spectral and texture features and a decision fusion mechanism to enhance the image properties, thus improving classification accuracy. Two hyperspectral datasets, originated from proximal hyperspectral systems, were used in the evaluation and significant improvements in classification accuracy achieved.

Spatial-spectral identification of abnormal leukocytes based on microscopic hyperspectral imaging technology

“…To gain insightful information of surface properties, the adjacent pixel intensities need to be analyzed or modeled. 28 The gray-level co-occurrence matrix (GLCM) studies the spatial relations in pairs of pixels. 29 Co-occurrence matrices relate the relative frequencies P ði; jjd; Þ, represent the number of occurrences of a pixel valued i which is separated from another pixel valued j at a distance d in the direction .…”

Spectral-spatial neural network classification of hyperspectral vegetation images

“…[8] To increase classification accuracy, a textural feature can be additionally considered. Textural features inherent in each type of vegetation can significantly increase the accuracy of classification [9]. Neural network-based algorithms can be used to analyze the texture.…”

Effective superpixel sparse representation classification method with multiple features and L0 smoothing for hyperspectral images

“…The abundant spectral–spatial information improves the ability of HSIs to identify different types of land cover. In recent years, HSI classification has been applied in precision agriculture, 1 , 2 mineralogy, 3 marine exploration, 4 object detection, 5 and other fields. Since HSIs have noise, redundant information between adjacent bands of HSIs, and the Hughes phenomenon caused by high dimensions, researchers have been dedicating their efforts to developing methods to improve the classification accuracy of HSIs.…”