Combined kernel-based BDT-SMO classification of hyperspectral fused images.
Ontology highlight
ABSTRACT:
SUBMITTER: Huang F
PROVIDER: S-EPMC4163336 | biostudies-other | 2014
REPOSITORIES: biostudies-other
Similar Datasets
Project description:Soyabean is an incredibly significant component of Chinese agricultural product, and categorizing soyabean seeds allows for a better understanding of the features, attributes, and applications of many species of soyabean. This enables farmers to choose appropriate seeds for sowing in order to increase production and quality. As a result, this thesis provides a method for classifying soybean seeds that uses hyperspectral RGB picture reconstruction. Firstly, hyperspectral images of seven varieties of soybean, H1, H2, H3, H4, H5, H6 and H7, were collected by hyperspectral imager, and by using the principle of the three base colours, the R, G and B bands which have more characteristic information are selected to reconstruct the images with different texture and colour characteristics to generate a new dataset for seed segmentation, and finally, a comparison is made with the classification effect of the seven models. The experimental results in ResNet34 show that the classification accuracy of the dataset before and after RGB reconstruction increases from 88.87% to 91.75%, demonstrating that RGB image reconstruction can strengthen image features; ResNet18, ResNet34, ResNet50, ResNet101, CBAM-ResNet34, SENet-ResNet34, and SENet-ResNet34-DCN models have classification accuracies of 72.25%, 91.75%, 89%, 88.48%, 92.28%, 92.80%, and 94.24%, respectively.SENet-ResNet34-DCN achieves the greatest classification accuracy results, with a model loss of roughly 0.3. The proposed SENet-ResNet34-DCN model is the most effective at classifying soybean seeds. By classifying and optimally selecting seed varieties, agricultural production can become more scientific, efficient, and sustainable, resulting in higher returns for farmers and contributing to global food security and sustainable development.
| S-EPMC11373792 | biostudies-literature
Project description:Background and objectiveParkinson's disease (PD) is a progressive neurological condition that manifests motor and non-motor symptoms. Early in the course of the disease, PD patients frequently experience vocal difficulties. In the beginning, preprocessing procedures were used with multi-focus image fusion to enhance the quality of input images. It is essential to diagnose and treat PD early to ensure that patients live healthy and productive lives.MethodsTremors, rigidity in the muscles, slow movement, difficulty balance, and other psychological symptoms are some of the disease's symptoms. One of the critical mechanisms supporting PD identification and assessment is the dynamics of handwritten records. Several machine-learning techniques have been researched for the early detection of this disease. Yet the main problem with most of these manual feature extraction methods is their poor performance and accuracy.ResultsThis cannot be acceptable when discovering such a chronic condition. For this purpose, a powerful deep learning model is suggested to help with the early diagnosis of Parkinson's disease. Therefore, we proposed MobileNetV3-based classification. To enhance the classification performances even more, the MobileNetV3-based approach was optimized by the Improved Dwarf Mongoose Optimization algorithm (IDMO).ConclusionThe Pyramid channel-based feature attention network (PCFAN) chooses the critical features. The efficiency of the approaches is tested using the PPMI and NTUA datasets. Our proposed approach obtains 99.34% accuracy, 98.53% sensitivity, 97.78% specificity, and 99.12% F-score compared to previous methods.
| S-EPMC10702996 | biostudies-literature
Project description:Surgery for brain cancer is a major problem in neurosurgery. The diffuse infiltration into the surrounding normal brain by these tumors makes their accurate identification by the naked eye difficult. Since surgery is the common treatment for brain cancer, an accurate radical resection of the tumor leads to improved survival rates for patients. However, the identification of the tumor boundaries during surgery is challenging. Hyperspectral imaging is a non-contact, non-ionizing and non-invasive technique suitable for medical diagnosis. This study presents the development of a novel classification method taking into account the spatial and spectral characteristics of the hyperspectral images to help neurosurgeons to accurately determine the tumor boundaries in surgical-time during the resection, avoiding excessive excision of normal tissue or unintentionally leaving residual tumor. The algorithm proposed in this study to approach an efficient solution consists of a hybrid framework that combines both supervised and unsupervised machine learning methods. Firstly, a supervised pixel-wise classification using a Support Vector Machine classifier is performed. The generated classification map is spatially homogenized using a one-band representation of the HS cube, employing the Fixed Reference t-Stochastic Neighbors Embedding dimensional reduction algorithm, and performing a K-Nearest Neighbors filtering. The information generated by the supervised stage is combined with a segmentation map obtained via unsupervised clustering employing a Hierarchical K-Means algorithm. The fusion is performed using a majority voting approach that associates each cluster with a certain class. To evaluate the proposed approach, five hyperspectral images of surface of the brain affected by glioblastoma tumor in vivo from five different patients have been used. The final classification maps obtained have been analyzed and validated by specialists. These preliminary results are promising, obtaining an accurate delineation of the tumor area.
| S-EPMC5858847 | biostudies-literature
Project description:This article presents a dataset of hyperspectral images of handwriting samples collected from 54 individuals. The purpose of the presented dataset is to further explore the use of hyperspectral imaging in document image analysis and to benchmark the performance of forensic analysis methods for hyperspectral document images. Each hyperspectral cube in the dataset has a spatial resolution of 512 × 650 pixels and contains 149 spectral channels in the spectral range of 478-901 nm. All the individuals have different personalities and have their writing patterns. The information of age and gender of each individual is collected. Each subject has written twenty-eight sentences using 12 different varieties of pens from different brands in blue color, each approximately 9 words or 33 characters long, all English alphabets in capital and small cases, digits from 0 to 9. The previous methods use synthetic mixed samples created by joining different parts of the images from the UWA WIHSI dataset.Each document consists of real mixed samples written withdifferent pens and by different writers with a variety of mixing ratios of inks and writers for forensic analysis.The standard A4 pages, each weighing 70 gs and manufactured by "AA" company, are used for data collection. The handwritten notes written by each subject with different pens are annotated in rectangular boxes. This dataset can be used for several tasks related to hyperspectral document image analysis and document forensic analysis including, handwritten optical character recognition, ink mismatch detection, writer identification at sentence, word, and character-level, handwriting-based gender classification, handwriting-based age prediction, handwritten word segmentation, and word generation. This dataset was designed and collected by the research team at the Artificial intelligence and Computer Vision Lab (iVision), Institute of Space Technology, Pakistan, and the hyperspectral images were acquired through imaging spectroscopy in the visible wavelength range at Wageningen University & Research, the Netherlands.
| S-EPMC8873541 | biostudies-literature
Project description:Hyperspectral imaging is a technique that provides rich chemical or compositional information not regularly available to traditional imaging modalities such as intensity imaging or color imaging based on the reflection, transmission, or emission of light. Analysis of hyperspectral imaging often relies on machine learning methods to extract information. Here, we present a new flexible architecture, the U-within-U-Net, that can perform classification, segmentation, and prediction of orthogonal imaging modalities on a variety of hyperspectral imaging techniques. Specifically, we demonstrate feature segmentation and classification on the Indian Pines hyperspectral dataset and simultaneous location prediction of multiple drugs in mass spectrometry imaging of rat liver tissue. We further demonstrate label-free fluorescence image prediction from hyperspectral stimulated Raman scattering microscopy images. The applicability of the U-within-U-Net architecture on diverse datasets with widely varying input and output dimensions and data sources suggest that it has great potential in advancing the use of hyperspectral imaging across many different application areas ranging from remote sensing, to medical imaging, to microscopy.
| S-EPMC8528004 | biostudies-literature
Project description:Background Tree crown extraction is an important research topic in forest resource monitoring. In particular, it is a prerequisite for disease detection and mapping the degree of damage caused by forest pests. Unmanned aerial vehicle (UAV)-based hyperspectral imaging is effective for surveying and monitoring forest health. This article proposes a spectral-spatial classification framework that uses UAV-based hyperspectral images and combines a support vector machine (SVM) with an edge-preserving filter (EPF) for completing classification more finely to automatically extract tree crowns damaged by Dendrolimus tabulaeformis Tsai et Liu (D. tabulaeformis) in Jianping county of Liaoning province, China. Results Experiments were conducted using UAV-based hyperspectral images, and the accuracy of the results was assessed using the mean structure similarity index (MSSIM), the overall accuracy (OA), kappa coefficient, and classification accuracy of damaged Pinus tabulaeformis. Optimized results showed that the OA of the spectral-spatial classification method can reach 93.17%, and the extraction accuracy of damaged tree crowns is 7.50–9.74% higher than that achieved using the traditional SVM classifier. Conclusion This study is one of only a few in which a UAV-based hyperspectral image has been used to extract tree crowns damaged by D. tabulaeformis. Moreover, the proposed classification method can effectively extract damaged tree crowns; hence, it can serve as a reference for future studies on both forest health monitoring and larger-scale forest pest and disease assessment.
| S-EPMC7547508 | biostudies-literature
Project description:The amount of information involved in hyperspectral imaging is large. Hyperspectral band selection is a popular method for reducing dimensionality. Several information based measures such as mutual information have been proposed to reduce information redundancy among spectral bands. Unfortunately, mutual information does not take into account the spatial dependency between adjacent pixels in images thus reducing its robustness as a similarity measure. In this paper, we propose a new band selection method based on spatial mutual information. As validation criteria, a supervised classification method using support vector machine (SVM) is used. Experimental results of the classification of hyperspectral datasets show that the proposed method can achieve more accurate results.
| S-EPMC4396555 | biostudies-other
Project description:Hyperspectral imaging (HSI) emerges as a non-destructive and rapid analytical tool for assessing food quality, safety, and authenticity. This work aims to investigate the potential of combining the spectral and spatial features of HSI data with the aid of deep learning approach for the pixel-wise classification of food products. We applied two strategies for extracting spatial-spectral features: (1) directly applying three-dimensional convolution neural network (3-D CNN) model; (2) first performing principal component analysis (PCA) and then developing 2-D CNN model from the first few PCs. These two methods were compared in terms of efficiency and accuracy, exemplified through two case studies, i.e., classification of four sweet products and differentiation between white stripe ("myocommata") and red muscle ("myotome") pixels on salmon fillets. Results showed that combining spectral-spatial features significantly enhanced the overall accuracy for sweet dataset, compared to partial least square discriminant analysis (PLSDA) and support vector machine (SVM). Results also demonstrated that spectral pre-processing techniques prior to CNN model development can enhance the classification performance. This work will open the door for more research in the area of practical applications in food industry.
| S-EPMC7570506 | biostudies-literature
Project description:BackgroundRecent studies have shown that hyperspectral imaging (HSI) combined with neural networks can detect colorectal cancer. Usually, different pre-processing techniques (e.g., wavelength selection and scaling, smoothing, denoising) are analyzed in detail to achieve a well-trained network. The impact of post-processing was studied less.MethodsWe tested the following methods: (1) Two pre-processing techniques (Standardization and Normalization), with (2) Two 3D-CNN models: Inception-based and RemoteSensing (RS)-based, with (3) Two post-processing algorithms based on median filter: one applies a median filter to a raw predictions map, the other applies the filter to the predictions map after adopting a discrimination threshold. These approaches were evaluated on a dataset that contains ex vivo hyperspectral (HS) colorectal cancer records of 56 patients.Results(1) Inception-based models perform better than RS-based, with the best results being 92% sensitivity and 94% specificity; (2) Inception-based models perform better with Normalization, RS-based with Standardization; (3) Our outcomes show that the post-processing step improves sensitivity and specificity by 6.6% in total. It was also found that both post-processing algorithms have the same effect, and this behavior was explained.ConclusionHSI combined with tissue classification algorithms is a promising diagnostic approach whose performance can be additionally improved by the application of the right combination of pre- and post-processing.
| S-EPMC10093280 | biostudies-literature
Project description:Cancer originates from the uncontrolled growth of healthy cells into a mass. Chromophores, such as hemoglobin and melanin, characterize skin spectral properties, allowing the classification of lesions into different etiologies. Hyperspectral imaging systems gather skin-reflected and transmitted light into several wavelength ranges of the electromagnetic spectrum, enabling potential skin-lesion differentiation through machine learning algorithms. Challenged by data availability and tiny inter and intra-tumoral variability, here we introduce a pipeline based on deep neural networks to diagnose hyperspectral skin cancer images, targeting a handheld device equipped with a low-power graphical processing unit for routine clinical testing. Enhanced by data augmentation, transfer learning, and hyperparameter tuning, the proposed architectures aim to meet and improve the well-known dermatologist-level detection performances concerning both benign-malignant and multiclass classification tasks, being able to diagnose hyperspectral data considering real-time constraints. Experiments show 87% sensitivity and 88% specificity for benign-malignant classification and specificity above 80% for the multiclass scenario. AUC measurements suggest classification performance improvement above 90% with adequate thresholding. Concerning binary segmentation, we measured skin DICE and IOU higher than 90%. We estimated 1.21 s, at most, consuming 5 Watts to segment the epidermal lesions with the U-Net++ architecture, meeting the imposed time limit. Hence, we can diagnose hyperspectral epidermal data assuming real-time constraints.
| S-EPMC9571453 | biostudies-literature