Fisher Linear Discriminant Analysis Research Articles

Domain Adaptation via Bregman divergence minimization

In recent years, Fisher linear discriminant analysis (FLDA) based classifi-cation is one of the most successful approaches and has been shown effectiveperformance in different classification tasks. However, when the learning data(source domain) have a different distribution against the testing data (tar-get domain), the FLDA-based models may not be optimal, and the perfor-mance will be degraded, dramatically. To face this problem, in this paper, wepropose an optimal domain adaptation via Bregman divergence minimization(DAB) approach, in which the discriminative features of source and target do-mains are simultaneously learned via domain invariant representation. DABis designed based on the constraints of FLDA, with the objective to adaptthe coupled marginal and conditional distribution mismatches with Breg-man divergence minimization. Thus, the resulting representation can havewell functionality like FLDA and simultaneously have better discriminationability. Moreover, our proposed approach can be easily kernelized to dealwith nonlinear tasks. Extensive experiments on various benchmark datasetsshow that our DAB can effectively deal with the cross domain divergence andoutperforms several state-of-the-art domain adaptation approaches on cross-distribution domains.

Assessment of rice seed vigour using selected frequencies of electrical impedance spectroscopy

The electrical impedance spectra of fresh and naturally aging rice seeds were measured to develop a non-destructive method for measuring the vigour of such seeds. To this end, three frequencies were selected using the coefficient of variation method within various frequency domains. A prediction model was then established using Fisher linear discriminant analysis, with the impedance, phase angle, and three-dimensional size of the seeds at the selected frequencies as independent variables and fresh and aging seeds as dependent variables. The water states and micromorphology of the seeds were subsequently evaluated to reveal the influence mechanism of impedance and vigour. The results showed that the impedance parameters of fresh and aging seeds were distinguishable, and the difference was evident in the low-frequency region. Moreover, the discriminant analysis model showed good predictive ability; the accuracies of the discriminant functions, Y 1 and Y 5 , were 90%. The impedance variables had a significant influence on the accuracy of the model, and the prediction accuracies for the fresh and aging seeds were improved by 24% and 14%, respectively, through introduction of the independent variables. Additionally, changes in the vigour, water state, and impedance were observed to be interrelated. The moisture content of the aging seeds was mainly affected by the content of bound water, and the ratio of bound water was inversely proportional to the impedance value. The variation in the impedance of the aging seeds was deemed to result from the combined effect of water state and micromorphology. • Electrical impedance spectroscopy parameters can reveal vigor level of rice seeds. • Moisture content of aging seeds is mainly affected by the content of bound water. • Ratio of bound water is inversely proportional to the impedance value. • Impedance variation combined effect of water state and micromorphology.

Nucleosides-based identification model for Fritillariae Cirrhosae Bulbus

A high performance liquid chromatography( HPLC) method was established for the fast,and precise determination of ten nucleosides in Fritillariae Cirrhosae Bulbus and its counterfeits. Then multivariate statistical analyses,such as clustering analysis,principal component analysis( PCA),and Fisher' s linear discriminant analysis( LDA),were conducted to establish a discriminant function model for an integrated analysis. The results indicated that data acquisition time of a single sample was shortened within 16 min by the HPLC method. In the range of 5-1 000 mg·kg~(-1),the mass concentrations of all nucleosides exhibited good linear relationships with the corresponding peak areas( R2> 0. 999). The spiked recoveries were in the range of 93. 83%-108. 9% with RSDs of0. 12%-1. 3%( n = 5). The limit of quantitation( LOQ) was 0. 98-4. 13 mg·kg~(-1). As revealed by the clustering analysis,Fritillariae Cirrhosae Bulbus and the counterfeits could be discriminated into two clusters based on the content of nucleosides. Fisher's LDA could achieve this discrimination,while PCA dimension reduction failed. The accuracy of the discriminant function model established on the screened characteristic indicators reached 97. 5%. The present study proposed a new identification method of Fritillariae Cirrhosae Bulbus with one-dimensional indicators,which is simple,accurate,and reliable. It can provide a scientific basis for further optimizing the identification techniques for Fritillariae Cirrhosae Bulbus and inspiration for quality control strategy development of Chinese medicinal materials.

The Association of Hiatal Dimensions and Urethral Mobility With Stress Urinary Incontinence.

To investigate the association of hiatal area (HA), bladder neck mobility, and urethral mobility during the cough stress test (CST) with stress urinary incontinence (SUI). This was a prospective study of 110 continent and 190 incontinent women using transperineal ultrasound. HA, bladder neck mobility, and the mobility of six points along the urethra (Vectors 1-6) were measured. The cohort was randomly divided at a ratio of 2:1 into a training cohort and a validation cohort. The correlations of HA with bladder neck mobility, urethral mobility, and SUI were tested. The predictive model was yielded by fisher linear discriminant analysis and receiver operating characteristics to assess the parameters' ability to predict SUI. Valid data were collected from 177 incontinent women and 105 continent women. Significant differences were identified in HA, body mass index (BMI), funneling, bladder neck mobility, and Vectors 1-6 between them. HA was positively correlated to bladder neck mobility. In the training cohort, bladder neck position on Valsalva, Vectors 3, and BMI had the area under curves of 0.74, 0.69, and 0.66 (all P< 0.001); Funneling and Vector 3 had odds ratios of 18.96 and 3.65 (all P< 0.001), for predicting SUI. The predictive model incorporating funneling, Vectors 3, and BMI provided the best performance in predicting SUI in both cohorts. The larger the HA was, the higher the bladder neck mobility. However, it was mid-urethral mobility rather than bladder neck mobility that performed best at predicting SUI.

Tracing the Origins of Blueberries by Their Mineral Element Contents and 87Sr/86Sr Ratios.

This study investigated the feasibility of using the mineral element contents and 87Sr/86Sr ratios of blueberries to trace their origins. The contents of 28 mineral elements, including K, Al, and Mg, were determined in 104 blueberry samples from three blueberry-producing areas in Guizhou Province, China. Also determined were both the 87Sr/86Sr ratios in 48 blueberry samples as well as the type of soil in which the blueberries were grown. Cluster analysis of 87Sr/86Sr ratios, stepwise discriminant analysis of mineral element contents, and combined discriminant analyses of 87Sr/86Sr ratios and mineral element contents were done. The results show that ten elements (Ca, Cr, Cs, Mg, Mn, P, Rb, Sb, Th, and Y) were selected by linear discriminant analysis, which could be used to establish the provenance traceability model of blueberries in Guizhou. The original accuracy of linear discriminant analysis was 89.4%, and the accuracy of cross-validation was 83.6%. When 87Sr/86Sr ratios were used for tracing, the accuracies of both the original discrimination and the cross-validation were 81.3% as determined by linear discriminant analysis, and the accuracy rate of origin discrimination was 81.25% by cluster analysis. The results of combined discrimination were the best: the accuracy of the original discrimination was 95.8%, and the accuracy of cross-validation was 93.8%. Mineral element contents and 87Sr/86Sr ratios can be used to trace the origins of blueberries, and combining them can significantly improve the discrimination accuracy. Fisher linear discriminant analysis had the greatest accuracy followed by cluster analysis, while principal component analysis was relatively poor in the research of Guizhou blueberry origin traceability.

Measures of Scatter and Fisher Discriminant Analysis for Uncertain Data

Uncertain data objects are objects that can be characterized by either a probability density function (PDF) or with multiple points. Because of existing levels of uncertainty for uncertain data objects, the scatter of this type of objects might be very different than the scatter of certain data objects. Measures of scatter for uncertain objects have not been defined before. In this paper, we define covariance matrix, within scatter matrix, and between scatter matrix as the measures of scatter for uncertain data objects. Also, in this paper, we extend the idea of Fisher linear discriminant analysis for uncertain objects (UFLDA). We also develop kernel Fisher discriminant analysis for uncertain objects (UKFDA). The developed uncertain kernel Fisher discriminants are for two cases: 1) when the uncertain objects are given with PDF and 2) when the uncertain objects are given with multiple points. We compare the performance of our developed uncertain Fisher discriminants (UFLDA and UKFDA) with a few other methods in classification of uncertain data objects through several examples on both simulated and real-world data. We will show in the experiments that our developed uncertain Fisher discriminants outperform other methods in classifying uncertain data objects.

Estimating Fisher discriminant error in a linear integrator model of neural population activity

Decoding approaches provide a useful means of estimating the information contained in neuronal circuits. In this work, we analyze the expected classification error of a decoder based on Fisher linear discriminant analysis. We provide expressions that relate decoding error to the specific parameters of a population model that performs linear integration of sensory input. Results show conditions that lead to beneficial and detrimental effects of noise correlation on decoding. Further, the proposed framework sheds light on the contribution of neuronal noise, highlighting cases where, counter-intuitively, increased noise may lead to improved decoding performance. Finally, we examined the impact of dynamical parameters, including neuronal leak and integration time constant, on decoding. Overall, this work presents a fruitful approach to the study of decoding using a comprehensive theoretical framework that merges dynamical parameters with estimates of readout error.

Deep Learning Features for Discriminating Between Benign and Malignant Microcalcification Lesions

Accurate diagnosis of microcalcification (MC) lesions in mammograms as benign or malignant is a challenging clinical task. In this study we investigate the potential discriminative power of deep learning features in MC lesion diagnosis. We consider two types of deep learning networks, of which one is a convolutional neural network developed for MC detection and the other is a denoising autoencoder network. In the experiments, we evaluated both the separability between malignant and benign lesions and the classification performance of image features from these two networks using Fisher's linear discriminant analysis on a set of mammographic images. The results demonstrate that the deep learning features from the MC detection network are most discriminative for classification of MC lesions when compared to both features from the autoencoder network and traditional handcrafted texture features.

A comparative study of 14 state of art descriptors for face recognition

This research paper presents a comparative study between 14 state of art descriptors which includes Local Binary Pattern (LBP), Median Binary Pattern (MBP), 6 × 6 Multiscale Block LBP (6 × 6 MB-LBP), Local Neighborhood Difference Pattern (LNDP), Logically Connected-LBP (LC-LBP), Local Phase Quantization (LPQ), Compound LBP (CLBP), Horizontal Elliptical LBP (HELBP), Vertical Elliptical LBP (VELBP), ELBP, Neighborhood Intensity Based LBP (NI-LBP), Median Robust Extended LBP Based on NI (MRELBP-NI), Radial Difference-LBP (RD-LBP) and Transition LBP (tLBP). For all the descriptors the features are extracted globally and the dimensionality of the feature size is reduced by employing Principal Component Analysis (PCA) and Fishers Linear Discriminant Analysis (FLDA). Finally classification is performed by Support Vector Machines (SVMs) and Nearest Neighbor (NN). Experiments are performed on 8 challenging databases which covers all the major challenges such as pose variations, illumination variations, expression variations and occlusion changes. The 8 challenging databases includes ORL, GT, Faces94, MIT-CBCL, Yale, YB, EYB and SOF. Out of all the descriptors it is the performance of the CLBP descriptor which is most encouraging. On some occasions the MRELBP-NI descriptor also achieves good results. But all in all the CLBP descriptor achieves the best results. In addition to this Deep learning based descriptors are also discussed in the paper.

Semantic Recognition of Human-Object Interactions via Gaussian-Based Elliptical Modeling and Pixel-Level Labeling

Human-Object Interaction (HOI) recognition, due to its significance in many computer vision-based applications, requires in-depth and meaningful details from image sequences. Incorporating semantics in scene understanding has led to a deep understanding of human-centric actions. Therefore, in this research work, we propose a semantic HOI recognition system based on multi-vision sensors. In the proposed system, the de-noised RGB and depth images, via Bilateral Filtering (BLF), are segmented into multiple clusters using a Simple Linear Iterative Clustering (SLIC) algorithm. The skeleton is then extracted from segmented RGB and depth images via Euclidean Distance Transform (EDT). Human joints, extracted from the skeleton, provide the annotations for accurate pixel-level labeling. An elliptical human model is then generated via a Gaussian Mixture Model (GMM). A Conditional Random Field (CRF) model is trained to allocate a specific label to each pixel of different human body parts and an interaction object. Two semantic feature types that are extracted from each labeled body part of the human and labelled objects are: Fiducial points and 3D point cloud. Features descriptors are quantized using Fisher's Linear Discriminant Analysis (FLDA) and classified using K-ary Tree Hashing (KATH). In experimentation phase the recognition accuracy achieved with the Sports dataset is 92.88%, with the Sun Yat-Sen University (SYSU) 3D HOI dataset is 93.5% and with the Nanyang Technological University (NTU) RGB+D dataset it is 94.16%. The proposed system is validated via extensive experimentation and should be applicable to many computer-vision based applications such as healthcare monitoring, security systems and assisted living etc.

Discovery and Validation of Serum Autoantibodies Against Tumor-Associated Antigens as Biomarkers in Gastric Adenocarcinoma Based on the Focused Protein Arrays.

INTRODUCTION:Previous studies have demonstrated that autoantibodies against tumor-associated antigens (TAAs) in patients with cancer can be used as sensitive immunodiagnostic biomarkers for the detection of cancer. Most of these TAAs are involved in the tumorigenesis pathway. Cancer driver genes with intragenic mutations can promote tumorigenesis. This study aims to identify autoantibodies against TAAs encoded by cancer driver genes in sera as potential immunodiagnostic biomarkers for gastric adenocarcinoma (GAC).METHODS:Protein arrays based on cancer driver genes were customized for screening candidate TAAs in 100 GAC sera and 50 normal control (NC) sera. Autoantibodies against candidate TAAs were assessed by enzyme-linked immunosorbent assay in both training group (205 GAC sera and 205 NC sera) and independent validation group (126 GAC sera and 126 NC sera). Moreover, the immunodiagnostic models were respectively established and validated in the training group and validation group.RESULTS:A panel with 5 autoantibodies including anti-TP53, anti-COPB1, anti-GNAS, anti–serine/arginine-rich splicing factor 2, and anti-SMARCB1 was selected by the Fisher linear discriminant analysis model with an areas under receiver operating characteristic curve (AUC) of 0.928 (95% confidence interval [CI]: 0.888–0.967) in the training cohort and an AUC of 0.885 (95% CI: 0.852–0.918) in the validation cohort. Besides, the panel with 5 autoantibodies including anti-TP53, anti-COPB1, anti-GNAS, anti-PBRM1, and anti-ACVR1B which were selected by the binary logistic regression model showed an AUC of 0.885 (95% CI: 0.852–0.919) in the training cohort and 0.884 (95% CI: 0.842–0.925) in the validation cohort.DISCUSSION:Two panels which were selected in this study could boost the detection of anti-TAA autoantibodies in sera as biomarkers for the detection of GAC.

The Effect of SOA on An Asynchronous ERP and VEP-Based BCI

Our previous study established an asynchronous BCI system by using the oddball paradigm to simultaneously induce event-related potentials (ERPs) and visual evoked potentials (VEPs) (E-V BCIs). We found that stimulus onset asynchrony (SOA) is an important factor for performance since it significantly affects the ERP and VEP. Increasing the SOA increases the ERP, which improves the accuracy of detecting target stimuli. However, a larger SOA leads to a lower VEP frequency, which causes the VEP to have poor accuracy when discriminating between the brain states. How to balance the two potentials and accuracies is a problem. This study established eight SOAs from 100 ms to 375 ms that were composed of different interstimulus intervals and the same stimulus duration of 80 ms. We used a probability-based Fisher linear discriminant analysis (P-FLDA) classifier to calculate the classification accuracies of the ERP-based visual speller, VEP-based brain state discrimination, and E-V BCI. The results show that as the SOA increases, the amplitudes of N200 and P300 increase, and the accuracy also shows an increasing trend. However, the frequency of the VEP and the accuracies of state discrimination show downward trends. The change in accuracies of the E-V BCI system combining these two parts is nonlinear, and the SOA optimal value is 125 ms. The SOA of 125 ms yields the best accuracies of 95.83% and practical bit rate of 57.17 bits/min in the E-V BCI system, which provides a guideline for selecting the SOA to improve the performance.

CLASSIFICATION OF DYNAMIC EGG WEIGHTS USING FEATURE EXTRACTION METHODS

In this study, a feature vector is determined in order to classify chicken eggs into four different weight groups by using the dynamic weighing system and then the success rate of different classifiers in the process of weight classification are analyzed. The dynamic weighing system is made of three components; mechanic system, electronic control board, and software. Firstly, a data set is created on the basis of analogue egg weight data obtained from the dynamic weighing system. From the obtained data set, three different feature vectors are extracted by using Time-domain, Power Spectral Density (PSD) and Discrete Wavelet Transform (DWT)-based methods. The extracted feature vectors are then applied to Linear Bayes Normal Classifier, Fisher’s Linear Discriminant Analysis (FLDA), Support Vector Machine (SVM), Decision Tree (DT) and K-Nearest Neighborhood (k-NN) classifiers respectively and egg weight classes are determined. A five-fold cross validation is carried out in order to confidentially test the performance of classification. As can be seen from the experimental results, both feature vectors and classifiers are highly successful in determining the weight classes of eggs. It is observed that the most successful features are the entropy values of DWT with a classification rate of 97.01% for k-NN classifier.

An Efficient Gait Recognition Method for Known and Unknown Covariate Conditions

Gait is a unique non-invasive biometric form that can be utilized to effectively recognize persons, even when they prove to be uncooperative. Computer-aided gait recognition systems usually use image sequences without considering covariates like clothing and possessions of carrier bags whilst on the move. Similarly, in gait recognition, there may exist unknown covariate conditions that may affect the training and testing conditions for a given individual. Consequently, common techniques for gait recognition and measurement require a degree of intervention leading to the introduction of unknown covariate conditions, and hence this significantly limits the practical use of the present gait recognition and analysis systems. To overcome these key issues, we propose a method of gait analysis accounting for both known and unknown covariate conditions. For this purpose, we propose two methods, i.e., a Convolutional Neural Network (CNN) based gait recognition and a discriminative features-based classification method for unknown covariate conditions. The first method can handle known covariate conditions efficiently while the second method focuses on identifying and selecting unique covariate invariant features from the gallery and probe sequences. The feature set utilized here includes Local Binary Patterns (LBP), Histogram of Oriented Gradients (HOG), and Haralick texture features. Furthermore, we utilize the Fisher Linear Discriminant Analysis for dimensionality reduction and selecting the most discriminant features. Three classifiers, namely Random Forest, Support Vector Machine (SVM), and Multilayer Perceptron are used for gait recognition under strict unknown covariate conditions. We evaluated our results using CASIA and OUR-ISIR datasets for both clothing and speed variations. As a result, we report that on average we obtain an accuracy of 90.32% for the CASIA dataset with unknown covariates and similarly performed excellently on the ISIR dataset. Therefore, our proposed method outperforms existing methods for gait recognition under known and unknown covariate conditions.

A novel method for the nondestructive classification of different-age Citri Reticulatae Pericarpium based on data combination technique.

The quality of Citri Reticulatae Pericarpium (CRP) is closely correlated with the aging time. However, CRPs in different storage ages are similar in appearance, and the young CRP may be labeled as the aged one to obtain the excess profit by some unscrupulous traders. Most traditional analysis methods are laborious and time‐consuming, and they can hardly realize the nondestructive classification. In this paper, a novel method based on near‐infrared diffuse reflectance spectroscopy (NIRDRS) and data combination technique for the nondestructive classification of different‐age CRPs was proposed. The CRPs in different storage ages (5, 10, 15, 20, and 25 years) were measured. The near‐infrared spectra of outer skin and inner capsule were obtained. Principal component analysis (PCA), soft independent modeling of class analogy (SIMCA), and Fisher's linear discriminant analysis (FLD), with different data pretreatment methods, were used for the classification analysis. Data combination of the outer skin and inner capsule spectra was discussed for further improving the classification results. The results show that multiple sensors provide more useful and complementary information than a single sensor does for improving the prediction accuracy. With the help of data combination strategy, 100% prediction accuracy can be obtained with both second‐order derivative–FLD and continuous wavelet transform–multiplicative scatter correction–FLD methods.

Hazard differentiation embedded in the brain: A near-infrared spectroscopy-based study

Comprehending the neurophysiological mechanics of hazard recognition (HR) is crucial in image analysis and automated hazard inspection. Although previous studies have used brain–-computer interfaces (BCIs) to reveal brain activity patterns during HR, few have considered construction hazards. In this study, we identified salient prefrontal cortex (PFC) areas that that signify different recognized hazards. We performed an experiment consisting of multiple HR tasks in a laboratory setting and recorded the hemodynamic responses with near-infrared spectroscopy (NIRS). The Fisher score and linear discriminant analysis were used to classify hazardous situations. The results showed that the left PFC was more engaged in HR: specifically, the dorsolateral PFC for electricity and impact-related HR and the ventrolateral PFC for stabbing-related HR. Theoretically, the identified critical areas for hazard differentiation can be regarded as a neuropsychological basis for cognition. Practically, these results demonstrate the potential of NIRS-based BCIs for hazard inspections.

Adaptive Multi-Channel Event Segmentation and Feature Extraction for Monitoring Health Outcomes.

To develop a multi-channel device event segmentation and feature extraction algorithm that is robust to changes in data distribution. We introduce an adaptive transfer learning algorithm to classify and segment events from non-stationary multi-channel temporal data. Using a multivariate hidden Markov model (HMM) and Fisher's linear discriminant analysis (FLDA) the algorithm adaptively adjusts to shifts in distribution over time. The proposed algorithm is unsupervised and learns to label events without requiring a priori information about true event states. The procedure is illustrated on experimental data collected from a cohort in a human viral challenge (HVC) study, where certain subjects have disrupted wake and sleep patterns after exposure to an H1N1 influenza pathogen. Simulations establish that the proposed adaptive algorithm significantly outperforms other event classification methods. When applied to early time points in the HVC data, the algorithm extracts sleep/wake features that are predictive of both infection and infection onset time. The proposed transfer learning event segmentation method is robust to temporal shifts in data distribution and can be used to produce highly discriminative event-labeled features for health monitoring. Our integrated multisensor signal processing and transfer learning method is applicable to many ambulatory monitoring applications.

Automated Classification of Glaucoma Stages Using Flexible Analytic Wavelet Transform From Retinal Fundus Images

Glaucoma is one of the leading causes of irreversible vision loss, it progresses gradually without easily noticeable symptoms. The detection of glaucoma in the early stage is crucial as it may help to decelerate the progress. The traditional instrument methods are manual, time-consuming and less accurate. Hence, the automated diagnosis of glaucoma is needed for detection of glaucoma in the early stage with high accuracy. The flexible analytic wavelet transform (FAWT) based novel method has been proposed for the classification of glaucoma stages. In the proposed method, FAWT has been used to decompose the preprocessed images into various sub-band images. Then, ReliefF and sequential box-counting (SBC) algorithms are applied to extract the various entropies and fractal dimension (FD) based features, respectively. Further, the extracted feature values are ranked using Fisher's linear discriminant analysis (LDA) dimensionally reduction. Finally, the higher rank features have been used for the classification of glaucoma stages using least squares-support vector machine (LS-SVM) classifier. The proposed method has been evaluated on publicly available large and diverse glaucoma database. The classification accuracy of the proposed method is 93.40% using tenfold cross-validation. The proposed method has demonstrated better performance for glaucoma classification as compared to the existing methods. The proposed method is ready to help the ophthalmologist in their daily screening for glaucoma detection.

Association of Face and Facial Components Based on CNN and Transfer Subspace Learning for Forensics Applications

Face and facial components are the most significant pieces of evidence used in forensic applications. Component-based recognition and age-invariant face recognition are the key factors in forensic face recognition. A major challenge during a forensic investigation is the association of facial components with a relevant face and this paper addresses this concern using the proposed transfer learning methodology. Research has been carried out treating facial components and the face as individual and separate entities. However, they share common auxiliary information which is exploited to bring the face and facial components in a common subspace in the proposed method. This facilitates transfer of knowledge gained from a face and classification of facial components of a person using the same subspace. Stability and invariant features of the facial components catalyze a solution for the problem of recognition. Convolutional neural network (CNN) is used to extract the features of the biometrics and then they are given to a regularization framework for reducing the probability distribution difference between them. The system is trained using the face and facial components and then tested using either the full-face or individual facial components. For the transfer, the proposed Fisher linear discriminant analysis and locality preserving projection, a convolutional neural network-based algorithm gives 91% and 90% accuracy, respectively, which outperforms the Histogram of Gradient and Gabor methods for predicting an association.

Wavelength weightings in machine learning for ovine joint tissue differentiation using diffuse reflectance spectroscopy (DRS).

Objective: To investigate the DRS of ovine joint tissue to determine the optimal optical wavelengths for tissue differentiation and relate these wavelengths to the biomolecular composition of tissues. In this study, we combine machine learning with DRS for tissue classification and then look further at the weighting matrix of the classifier to further understand the key differentiating features. Methods: Supervised machine learning was used to analyse DRS data. After normalising the data, dimension reduction was achieved through multiclass Fisher's linear discriminant analysis (Multiclass FLDA) and classified with linear discriminant analysis (LDA). The classifier was first run with all the tissue types and the wavelength range 190 nm - 1081 nm. We analysed the weighting matrix of the classifier and then ran the classifier again, the first time using the ten highest weighted wavelengths and the second using only the single highest. Our method was applied to a dataset containing ovine joint tissue including cartilage, cortical and subchondral bone, fat, ligament, meniscus, and muscle. Results: It achieved a classification accuracy of 100% using the wavelength 190 nm - 1081 nm (2048 attributes) with an accuracy of 90% being present for 10 attributes with the exception of those with comparable compositions such as ligament and meniscus. An accuracy greater than 70% was achieved using a single wavelength, with the same exceptions. Conclusion: Multiclass FLDA combined with LDA is a viable technique for tissue identification from DRS data. The majority of differentiating features existed within the wavelength ranges 370-470 and 800-1010 nm. Focusing on key spectral regions means that a spectrometer with a narrower range can potentially be used, with less computational power needed for subsequent analysis.