Structure Tensor and Guided Filtering-Based Algorithm for Hyperspectral Anomaly Detection

Quality inspection of fruits and vegetables linked to food safety monitoring and quality control. Traditional chemical analysis and physical measurement techniques are reliable, they are also time-consuming, costly, and susceptible to environmental and sample changes. Hyperspectral imaging technology combined with deep learning methods can effectively overcome these problems. Compared with human evaluation, automated inspection improves inspection efficiency, reduces subjective error, and promotes the intelligent and precise fruit and vegetable quality inspection. This paper reviews reports on the application of hyperspectral imaging technology combined to deep learning methods in various aspects of fruits and vegetables quality assessment. In addition, the latest applications of these technologies in the fields of fruit and vegetable safety, internal quality, and external quality inspection are reviewed, and the challenges and future development directions of hyperspectral imaging technology combined with deep learning in this field are prospected. Hyperspectral imaging combined with deep learning has shown significant advantages in fruit and vegetable quality inspection, especially in improving inspection accuracy and efficiency. Future research should focus on reducing costs, optimizing equipment, personalizing feature extraction, and model generalizability. In addition, the development of lightweight models and the balance of accuracy, the enhancement of the database and the importance of quantitative research should also be brought to attention. These efforts will promote the wide application of hyperspectral imaging technology in fruit and vegetable inspection, improve its practicability in the actual production environment, and bring important progress for food safety and quality management.

The Reed–Xiaoli (RX) algorithm has been widely used as an anomaly detector for hyperspectral images. Recently, kernel RX (KRX) has been proven to yield high performance in anomaly detection and change detection. In this paper, we present a generalization of the KRX algorithm. The novel algorithm is called cluster KRX (CKRX), which becomes KRX under certain conditions. The key idea is to group background pixels into clusters and then apply a fast eigendecomposition algorithm to generate the anomaly detection index. Both global and local versions of CKRX have been implemented. Application to anomaly detection using actual hyperspectral images is included. In addition to anomaly detection, the CKRX algorithm has been integrated with other prediction algorithms for change detection. Spatially registered visible and near-infrared hyperspectral images collected from a tower-based geometry have been used in the anomaly and change detection studies. Receiver operating characteristics curves and actual computation times were used to compare different algorithms. It was demonstrated that CKRX has comparable detection performance as KRX, but with much lower computational requirements.

Hyperspectral imaging is endowed with outstanding discriminability between different land types by its comprehensive sensing of the spectrum, thus favored applying to anomaly detection. However, blurring effect, as a critical cause for quality deterioration of hyperspectral imaging, has been omitted by previous hyperspectral anomaly detection models. On one hand, given that anomalies are sparsely distributed in nature, such blurring effect entangling neighboring pixels severely weighs those detection models down. On the other hand, abnormal objects jeopardize the low-dimensional structure of the image, thus deblurring those images with anomalies is more challenging than normal ones. Hence, it is of much significance to investigate anomaly detection using blurred hyperspectral images. To this end, this paper proposes a generalized non-convex surrogated tensor framework that is able to perform anomaly detection robustly to blurring effects on hyperspectral images. The proposed framework is featured to be a unified paradigm which guarantees convergence for a broad class of non-convex surrogates. Through treating the spatial and spectral low-rankness adaptively via Block Term Decomposition, the unevenness in the multi-linear low-rankness of hyperspectral image is comprehensively considered, which together with the non-convex surrogates results in a tighter modeling of the low-dimensional prior of hyperspectral images. Extensive experiments demonstrate the superiority of the proposed method compared with the state-of-the-art methods on both hyperspectral image deblurring and anomaly detection.

Paddy seed viability prediction based on feature fusion of color and hyperspectral image with multivariate analysis

Whether or not a hyperspectral anomaly detector is effective is determined by two crucial issues, anomaly detectability and background suppressibility (BS), both of which are very closely related to two factors, the datasets used for a selected hyperspectral anomaly detector and detection measures used for its performance evaluation. This paper explores how anomaly detectability and BS play key roles in hyperspectral anomaly detection (HAD). To address these two issues, we investigate three key elements attributed to HAD. One is a selected hyperspectral anomaly detector, and another is the datasets used for experiments. The third one is the detection measures used to evaluate the effectiveness of a hyperspectral anomaly detector. As for hyperspectral anomaly detectors, twelve commonly used anomaly detectors were evaluated and compared. To address the appropriate use of datasets for HAD, seven popular and widely used datasets were studied for HAD. As for the third issue, the traditional area under a receiver operating characteristic (ROC) curve of detection probability—PD versus false alarm probability, PF, (AUC(D,F))—was extended to 3D ROC analysis where a 3D ROC curve was developed to generate three 2D ROC curves from which eight detection measures could be derived to evaluate HAD in all round aspects, including anomaly detectability, BS and joint anomaly detectability and BS. Qualitative analysis showed that many works reported in the literature which claimed that their developed hyperspectral anomaly detectors performed better than other anomaly detectors are actually not true because they overlooked these two issues. Specifically, a comprehensive study via extensive experiments demonstrated that these 3D ROC curve-derived detection measures can be further used to address the various characterizations of different data scenes and also to provide explanations as to why certain data scenes are not suitable for HAD.

Collaborative representation-based detector (CRD), as the most representative anomaly detection method, has been widely applied in the field of hyperspectral anomaly detection (HAD). However, the sliding dual window of the original CRD introduces high computational complexity. Moreover, most HAD models only consider a single spectral or spatial feature of the hyperspectral image (HSI), which is unhelpful for improving detection accuracy. To solve these problems, in terms of speed and accuracy, we propose a novel anomaly detection approach, named Random Collective Representation-based Detector with Multiple Feature (RCRDMF). This method includes the following steps. This method first extract the different features include spectral feature, Gabor feature, extended multiattribute profile (EMAP) feature, and extended morphological profile (EMP) feature matrix from the HSI image, which enables us to improve the accuracy of HAD by combining the multiple spectral and spatial features. The ensemble and random collaborative representation detector (ERCRD) method is then applied, which can improve the anomaly detection speed. Finally, an adaptive weight approach is proposed to calculate the weight for each feature. Experimental results on six hyperspectral datasets demonstrate that the proposed approach has the superiority over accuracy and speed.

Fusion of hyperspectral and panchromatic images using structure tensor and matting model

Accurate classification is one of the most important prerequisites for hyperspectral applications and feature extraction is the key step of classification. Recently, deep learning models have been successfully used to extract the spectral-spatial features in hyperspectral images (HSIs). However, most deep learning-based classification methods are supervised and require sufficient samples to guarantee their performance. And the labeled samples in HSI are limited. To solve this problem, unsupervised feature extraction based on improved Wasserstein generative adversarial network (WGAN) is proposed in this paper. Further, in order to fully explore the spectral-spatial features in HSIs, a three-dimensional (3D) model is designed. Considering that its difficult for GANs to generate high dimension data, the dimension of HSI is reduced firstly by principle component analysis (PCA). Then, an improved WGAN is trained unsupervised with HSIs after dimensionality reduction to achieve the stable status. Finally, the discriminator of the trained improved WGAN is used as a feature extractor and followed by a classifier, which can be used to classify the HSIs. In order to evaluate the performance of the proposed method, PCA-based methods and GAN-based methods are compared in the experiment of a real-world HSI. Experimental results have shown that our proposed method gained more promising results which has great potential prospects in HSI classification.

Hyperspectral image (HSI) classification remains a challenging problem due to unique characteristics of HSI data (such as numerous bands and strong correlations in the spectral and spatial domains) and small sample size. To address such concerns, we propose a novel spectral–spatial feature extraction method for HSI classification by employing graph embedding and deep learning (DL) models. Since the conventional graph cannot capture the complex manifold relationship of HSI data, and there exist the observations of within-class variation as well as the similarity between different classes in the spectral domain, we construct the supervised within-class/between-class hypergraph (SWBH) to extract the spectral features of HSI. Since it is difficult for DL models to learn representative features for HSI data when the labeled training samples are limited, we propose the random zero settings to newly generate a large amount of labeled HSI samples for the training of convolutional neural network (CNN). The designed sample expanded CNN (SECNN) is used to extract the HSI spatial features. Thus, the spectral–spatial features of HSI can be learned by integrating the features extracted from SWBH and SECNN, respectively. Experiments on three real HSI datasets demonstrate higher classification accuracy of the proposed SWBH–SECNN method.

Lately, generative adversarial networks (GAN)-based methods have drawn extensive attention and achieved a promising performance in the field of hyperspectral anomaly detection (HAD) owing to GAN’s powerful data generation capability. However, without considering the background spatial features, most of these methods can not obtain a GAN with a strong background generation ability. Besides, they fail to address the hyperspectral image (HSI) redundant information disturbance problem in the anomaly detection part. To solve these issues, the unsupervised generative adversarial network with background spatial feature enhancement and irredundant pooling (BEGAIP) is proposed for HAD. To make better use of features, spatial and spectral features union extraction idea is also applied to the proposed model. To be specific, in spatial branch, a new background spatial feature enhancement way is proposed to get a data set containing relatively pure background information to train GAN and reconstruct a more vivid background image. In a spectral branch, irredundant pooling (IP) is invented to remove redundant information, which can also enhance the background spectral feature. Finally, the features obtained from the spectral and spatial branch are combined for HAD. The experimental results conducted on several HSI data sets display that the model proposed acquire a better performance than other relevant algorithms.

In Chap. 5, a particular subtarget detection technique in active hyperspectral target detection, called constrained energy minimization (CEM), was developed for its real-time and causal implementation. Rather than CEM, this chapter focuses on passive hyperspectral target detection and investigates a commonly used passive target detection technique, anomaly detection (AD), especially for real-time and causal processing capabilities that are developed for CEM in Chap. 5 and will also be derived in a similar manner for AD. To this end, this chapter can be considered a companion chapter of Chap. 5. Technically speaking, passive target detection must be carried out without any prior knowledge, specifically, a complete lack of availability of target knowledge. Owing to the nature of hyperspectral imaging sensors, which can uncover many unknown material substances, passive hyperspectral target detection is a major task in hyperspectral image analysis and has been studied extensively in the literature. Applications of passive target detection include surveillance and monitoring, where no knowledge is required a priori. Of particular interest in passive target detection is AD, which is generally performed in a completely blind environment. To suppress unknown backgrounds, AD makes use of the global sample correlation/covariance matrix R/K, referred to as R/K-AD, so as to enhance its detectability. In general, anomalies appear unexpectedly and cannot be detected by visual inspection. Most importantly, they vary with time and data sample vectors. Accordingly, developing real-time processing for R/K-AD on a timely basis sample by sample is crucial and critical in many real-world applications, for example, abnormalities in agriculture and forestry, environmental monitoring, combat vehicles on a battlefield, drug trafficking in law enforcement, and food inspection. As noted in Chap. 5, this real-time process requires that causality be included in an algorithm’s design. In analogy with real-time CEM, the concept of a causal sample correlation matrix (CSCRM), introduced in Chap. 5, is also applicable to AD for this purpose. However, in a broader sense, CSCRM is expanded into a causal sample correlation/covariance matrix (CSCRM/CSCVM) to replace R/K with CSCRM/CSCVM specified by R(r n )/K(r n ), respectively, where r n is the current data sample vector being processed. The capability for real-time processing is then derived from such a CSCRM/CSCVM because CSCRM/CSCVM varies with data sample vectors. Thus, a causal processing of K-AD/R-AD requires repeatedly calculating inverses of such sample-varying CSCRM/CSCVM and must be processed in a causal manner sample by sample, referred to as causal K-AD/R-AD (CR-K-AD/CR-AD). To further reduce computational complexity and computer processing time for CK-AD/CR-AD, the notion of innovation in a Kalman filter developed in Chap. 3 is once again used to derive recursive causal update equations to calculate CSCRM/CSCVM, which leads to recursive versions of K-AD/R-AD, to be called recursive CK-AD/CR-AD (R-CK-AD/R-CR-AD). In particular, when R-CK-AD/R-CR-AD is realized by real-time processing, it is referred to as RT CK-AD/RT-CR-AD in this chapter.

For accurate classification of remote sensing data, Hyperspectral Images (HSI) have become very popular. It can capture the reflected electromagnetic spectrum from the object in several contiguous spectral bands. But processing of hundreds of bands is computationally expensive and also it contains several noisy and redundant bands. Often the water absorption bands are manually removed by the researchers in advance. In this work, a histogram based automatic noisy band removal algorithm is developed for the HSI. This algorithm can be used as a preprocessing step prior to hyperspectral image classification. At first, by using the histogram information, noisy bands are removed. Next, after obtaining the desired number of non-noisy bands, a Gaussian Filter is applied on obtained bands to extract spatial-spectral features. Finally, to evaluate the algorithm, classification is performed using a SVM classifier. For experimental validation of results, Indian Pines and Salinas datasets are used. The obtained result clearly reveals the effectiveness of the proposed automatic noisy band removal algorithm.

To overcome the difficulty of automating and intelligently classifying the ground features in remote-sensing hyperspectral images, machine learning methods are gradually introduced into the process of remote-sensing imaging. First, the PaviaU, Botswana, and Cuprite hyperspectral datasets are selected as research subjects in this study, and the objective is to process remote-sensing hyperspectral images via machine learning to realize the automatic and intelligent classification of features. Then, the basic principles of the support vector machine (SVM) and extreme learning machine (ELM) classification algorithms are introduced, and they are applied to the datasets. Next, by adjusting the parameter estimates using a restricted Boltzmann machine (RBM), a new terrain classification model of hyperspectral images that is based on a deep belief network (DBN) is constructed. Next, the SVM, ELM, and DBN classification algorithms for hyperspectral image terrain classification are analysed and compared in terms of accuracy and consistency. The results demonstrate that the average detection accuracies of ELM on the three datasets are 89.54%, 96.14%, and 96.28%, and the Kappa coefficient values are 0.832, 0.963, and 0.924; the average detection accuracies of SVM are 88.90%, 92.11%, and 91.68%, and the Kappa coefficient values are 0.768, 0.913, and 0.944; the average detection accuracies of the DBN classification model are 92.36%, 97.31%, and 98.84%, and the Kappa coefficient values are 0.883, 0.944, and 0.972. The results also demonstrate that the classification accuracy of the DBN algorithm exceeds those of the previous two methods because it fully utilizes the spatial and spectral information of hyperspectral remote-sensing images. In summary, the DBN algorithm that is proposed in this study has high application value in object classification for remote-sensing hyperspectral images.

Recently, broad learning system (BLS) have demonstrated excellent performance in hyperspectral images (HSI) classification. However, due to the complex geometric structure and spatial layout of HSI, the linear sparse features in broad learning system are difficult to fully represent hyperspectral data. In addition, the features learned by broad learning system lack more effective discriminative ability, which leads to the limited expressive ability of features. To address the issues, we propose a graph convolutional enhanced discriminative broad learning system (GCDBLS) for HSI Classification. GCDBLS aggregates the node information in the adjacency graph through graph convolution, and then learns the context relationship, so as to obtain rich nonlinear spatial spectral features in hyperspectral images; In order to extract more discriminative hyperspectral image features, GCDBLS introduces the concept of local intra-class scatter and local inter-class scatter. By minimizing the local intra-class feature distance and maximizing the local inter-class feature distance, GCDBLS can improve the discrimination ability of BLS extracted features. On three HSI datasets, the experiments compared with the latest classification methods show that the proposed method achieves good results, and improves the classification performance of hyperspectral images.

Anomaly detection in hyperspectral imaging is crucial for remote sensing, driving the development of numerous algorithms. However, systematic studies reveal a dichotomy where algorithms generally excel at either detecting anomalies in specific datasets or generalizing across heterogeneous datasets (i.e., lack adaptability). A key source of this dichotomy may center on the singular and like biases frequently employed by existing algorithms. Current research lacks experimentation into how integrating insights from diverse biases might counteract problems in singularly biased approaches. Addressing this gap, we propose stacking-based ensemble learning for hyperspectral anomaly detection (SELHAD). SELHAD introduces the integration of hyperspectral anomaly detection algorithms with diverse biases (e.g., Gaussian, density, partition) into a singular ensemble learning model and learns the factor to which each bias should contribute so anomaly detection performance is optimized. Additionally, it introduces bootstrapping strategies into hyperspectral anomaly detection algorithms to further increase robustness. We focused on five representative algorithms embodying common biases in hyperspectral anomaly detection and demonstrated how they result in the previously highlighted dichotomy. Subsequently, we demonstrated how SELHAD learns the interplay between these biases, enabling their collaborative utilization. In doing so, SELHAD transcends the limitations inherent in individual biases, thereby alleviating the dichotomy and advancing toward more adaptable solutions.