Hyperspectral Image Classification across Different Datasets: A Generalization to Unseen Categories

Abstract

With the rapid developments of hyperspectral imaging, the cost of collecting hyperspectral data has been lower, while the demand for reliable and detailed hyperspectral annotations has been much more substantial. However, limited by the difficulties of labelling annotations, most existing hyperspectral image (HSI) classification methods are trained and evaluated on a single hyperspectral data cube. It brings two significant challenges. On the one hand, many algorithms have reached a nearly perfect classification accuracy, but their trained models are hard to generalize to other datasets. On the other hand, since different hyperspectral datasets are usually not collected in the same scene, different datasets will contain different classes. To address these issues, in this paper, we propose a new paradigm for HSI classification, which is training and evaluating separately across different hyperspectral datasets. It is of great help to labelling hyperspectral data. However, it has rarely been studied in the hyperspectral community. In this work, we utilize a three-phase scheme, including feature embedding, feature mapping, and label reasoning. More specifically, we select a pair of datasets acquired by the same hyperspectral sensor, and the classifier learns from one dataset and then evaluated it on the other. Inspired by the latest advances in zero-shot learning, we introduce label semantic representation to establish associations between seen categories in the training set and unseen categories in the testing set. Extensive experiments on two pairs of datasets with different comparative methods have shown the effectiveness and potential of zero-shot learning in HSI classification.

Entropy as a High-Level Feature for XAI-Based Early Plant Stress Detection

“…To determine N , first assume a separate thermal IR channel (TIR), a series of indices, which can be calculated using HSI channel images commonly used in smart agriculture: NDVI, GNDVI, GCL, SIPI, and GI [ 20 ]. In addition, consider the capabilities of the 3 visible HSI channels (R 630 , G 550 , and B 480 ) as analogs of the red, green, and blue channels of the regular color image.…”

“…In the construction of XAI-based classifier and regressor, which are simple and easily configurable as part of the user task, we followed the example described in [27], implementing the idea of [28]; however, we used the 'Backyard Dog' function as the main function of the network, which we implemented via SLP (Figure 1). To determine N, first assume a separate thermal IR channel (TIR), a series of indices, which can be calculated using HSI channel images commonly used in smart agriculture: NDVI, GNDVI, GCL, SIPI, and GI [20]. In addition, consider the capabilities of the 3 visible HSI channels (R630, G550, and B480) as analogs of the red, green, and blue channels of the regular color image.…”

“…An important condition for the successful application of computational experiments is the correct separation of plant pixels from the background under conditions of the changing state of the plant, soil, and other background objects. The plant mask can be To determine N, first assume a separate thermal IR channel (TIR), a series of indices, which can be calculated using HSI channel images commonly used in smart agriculture: NDVI, GNDVI, GCL, SIPI, and GI [20]. In addition, consider the capabilities of the 3 visible HSI channels (R 630 , G 550 , and B 480 ) as analogs of the red, green, and blue channels of the regular color image.…”

“…Of considerable interest in the field of XAI is the search for explainable features that are simple and fundamental enough to remain explainable for a wide class of objects and phenomena. Our interest in such explainable features was inspired by a recent article [ 20 ] on the zero-shot learning method based on a high-level feature vector. Such a vector ensures the minimization of the dataset required for training, as well as the transfer of training results from one dataset to another, in which some of the categories are missing (or not visible).…”

Gabor Features Extraction and Land-Cover Classification of Urban Hyperspectral Images for Remote Sensing Applications

“…These images can be used for different applications, for example soil evaluation for crop health and protection [1], maritime traffic surveillance [2], detection of minerals presence [3] and urban characterisation [4]. Moreover, the acquisition of this kind of image has presented time and cost reduction in recent years due to the improvement in portable technologies and transmission platforms [5,6]. Therefore, these advances have led to the processing of a significant volume of data growing at an accelerated speed, which originates data processing challenges [7].…”