Discriminant Analysis Algorithm Research Articles

Detection of Water Stress in Khasi Mandarin Orange Plants from Volatile Organic Compound Emission Profile Implementing Electronic Nose

Plants in the absence of an innate immune system like animals and being immobile are regularly exposed to a host of stresses, ranging from biotic to abiotic stresses. In response to these, plants have developed a complicated response system like reprogramming gene expressions and emission of secondary metabolites as volatile organic compounds (VOCs) by its various tissues like roots, stems, leaves etc. These VOCs can be used as biomarkers for inspecting plants’ in situ health status. This paper address the usefulness of electronic nose (e-nose) system to sense the VOCs emitted by plants’ leaves to detect the stresses in it. Standard commercial electronic nose (e-nose) system Alfa Mos Fox 3000 has been used here to identify the stressed and non-stressed plants. Fifteen Mandarin orange plants were considered for the study and were subdivided into three categories. Each one was subjected to a different level of water stress. Leaf samples were collected for e-nose analyses from each plant of all three categories on the 15th day and 30th day of induction of water stresses. Dimensionality reduction techniques like kernel Principal Component Analysis (kPCA), Linear Discriminant Analysis (LDA) and classification algorithms like Support Vector Machines (SVC) and Multi-Layer Perceptron Classifier (MLPC) have been used to classify the three categories of plants. The scores obtained from these analyses reveals the feasibility of using an e nose system in discriminating plants based on the status of water stress in them. This paper analyses the applicability of e nose system in stress diagnosis of agricultural and horticultural crops, which would significantly help in controlling the irrigation regime.

DISCRIMINATION OF TOMATO PLANTS (SOLANUM LYCOPERSICUM) GROWN UNDER ANAEROBIC BAFFLED REACTOR EFFLUENT, NITRIFIED URINE CONCENTRATE AND COMMERCIAL HYDROPONIC FERTILIZER REGIMES USING MULTI-SOURCE SATELLITE

Abstract. We evaluate the detection and discriminative strength of three different satellite spectral settings, namely, HyspIRI, the forthcoming Landsat 9 and Sentinel 2-MSI, in mapping tomato (Solanum lycopersicum) plants grown under hydroponic system using humanexcreta derived materials (HEDM), namely, anaerobic baffled reactor (ABR) effluent and nitrified urine concentrate (NUC) and commercial hydroponic fertilizer mix (CHFM) as nutrient sources. Partial least squares – discriminant analysis (PLS-DA) and discriminant analysis (DA) were applied to discriminate tomatoes grown under these different nutrient sources. Results of this study showed that spectral settings of HyspIRI sensor can better discriminate tomatoes grown under different fertilizer regimes when compared to Landsat 9 OLI and Sentinel-2 MSI spectral configurations. For instance, based on DA algorithm, HyspIRI exhibited high overall accuracy of 0.99 and a kappa statistic of 0.99 whereas Landsat OLI and Sentinel-2 MSI exhibited over accuracies of 0.94 and 0.95 as well as kappa statistics of 0.79 and 0.85, respectively. Further, the performance of DA was significantly different (α = 0.05) from that of PLS-DA based on the MaNemar tests. Overall, the performance of HyspIRI, Landsat 9 OLI-2 and Sentinel-2 MSI data seem to bring new opportunities for crop monitoring at farm scale.

USABILITY, STRENGTH AND PRACTICALITY OF THE UPCOMING HYSPIRI IN DETECTING MAIZE GRAY LEAFY SPOT IN RELATION TO SENTINEL-2 MSI, VENμS AND LANDSAT 8 OLI SPECTRAL BAND-SETTINGS

Abstract. Considering the high maize yield loses that are caused by diseases incidences as well as incomprehensive monitoring initiatives in the crop farming sector of agriculture, there is a need to come up with spatially explicit, cheap, fast and consistent approaches for monitoring as well as forecasting food crop diseases, such as maize gray leaf spot. This study, therefore, we sought to investigate the usability, strength and practicality of the forthcoming HyspIRI in detecting disease progression of Maize Gray leafy spot infections in relation to the Sentinel-2 MSI, Landsat 8 OLI spectral configurations. Maize Gray leafy spot disease progression that were discriminated based on partial least squares –discriminant analysis (PLS-DA) algorithm were (i) healthy, (ii) intermediate and (ii) severely infected maize crops. Comparatively, the results show that the HyspIRI’s simulated spectral settings slightly performed better than those of Sentinel-2 MSI, VENμS and Landsat 8 OLI sensor. HyspIRI exhibited an overall accuracy of 0.98 compared to 0.95, 0.93 and 0.89 exhibited by Sentinel-2 MSI, VENμS and Landsat 8 OLI sensor sensors, respectively. Further, the results showed that the visible section the red-edge and NIR covered by all the four sensors were the most influential spectral regions for discriminating different Maize Gray leafy spot infections. These findings underscore the added value and potential scientific breakthroughs likely to be brought about by the upcoming hyperspectral HyspIRI sensor in precision agriculture and forecasting of crop disease epidemics to ensure food security.

Maize seed classification using hyperspectral image coupled with multi-linear discriminant analysis

Seed purity is an important parameter for evaluating seed quality and can be effectively studied by seed classification. Hyperspectral images between 400 and 1000 nm were acquired for 1632 maize seeds (17 varieties) for classifying seed varieties. Fourteen features including a spectral feature and 13 imaging features (i.e., 5 first-order and 8 s-order textural features) were extracted from the hyperspectral image data. A multi-linear discriminant analysis (MLDA) algorithm was developed to select the optimal wavelength and transform/reduce the classification features to improve the acquisition and processing speed of the hyperspectral images. Least square support vector machine was used to develop classification models based on MLDA with spectral features, imaging features, and combination of spectral and imaging features. The effects of MLDA, uninformative variable elimination (UVE) coupled with linear discriminant analysis (LDA), and successive projection algorithm (SPA) coupled with LDA were adopted. Experimental results indicate that the combination feature based on the wavelength selection algorithm of MLDA yielded high classification accuracy under the same number of wavelengths (varying between 5 and 15). Meanwhile, the classification model based on MLDA feature transformation/reduction method achieved superior classification accuracy of 99.13% over SPA coupled with LDA (90.31%) and UVE coupled with LDA (94.17%) and improved by 2.74% relative to that of the mean spectrum of the full wavelength model. The proposed method can be used effectively for seed identification and classification.

Climate and soils at the Brazilian semiarid and the forest-Caatinga problem: new insights and implications for conservation

This study aimed to test two hypotheses: (i) on the Brazilian semiarid territory, the climate has greater weight as a driver of vegetation than the soil and; (ii) the arboreal Caatinga is a vegetation whose environmental attributes are similar to the Dry Forest, in terms of soil and climate attributes. We analyzed attributes of the superficial horizon of 156 standardized profiles distributed throughout the Brazilian semiarid region. Bioclimatic variables were obtained from the WorldClim platform and extracted to profiles location. The main vegetation types in the region were considered: Caatinga, arboreal Caatinga, Dry Forest and Cerrado. Variable selection was performed with hierarchical correlation dendrogram and recursive feature elimination algorithm. Linear Discriminant Analysis and Random Forest (RF) algorithm were used for modeling the edaphic and climate niche and predict the vegetation with the selected variables. Climate and soil, individually, were able to separate the vegetation, but the climate was no better predictor than the soil. Therefore, we reject the first hypothesis. However, the better prediction was attained with the combined use of soil and climate attributes. The parsimonious RF model had good performance, with Kappa 0.61 ± 0.10 and 70.9% ± 7.7% accuracy. The combination of soil and climate predictors resulted in better separation of vegetation in the Brazilian semiarid region. Soil attributes are key variables in large-scale biogeographic modeling. The so-called arboreal Caatinga is distributed over a wide edaphic and climatic range, with strong similarity to the Dry Forest distribution, confirmed by the great overlap in the multivariate space, which confirms the second hypothesis. The results point towards an urgent review of the Atlantic Forest Law. The environments where the arboreal Caatinga and the Dry Forest occur are very similar, so that the former may represent a degraded phase of the Atlantic Forest, currently without the due legal protection.

Improved LDA Dimension Reduction Based Behavior Learning with Commodity WiFi for Cyber-Physical Systems

In recent years, rapid development of sensing and computing has led to very large datasets. There is an urgent demand for innovative data analysis and processing techniques that are secure, privacy-protected and sustainable. In this article, taking human activities and interactions with Cyber-Physical Systems (CPS) into consideration, we propose a human behavior learning system based on Channel State Information (CSI) utilizing a series of algorithms for data analysis and processing. Aiming to recognize a set of gestures, our system is designed based on the observation that different gestures have different effects on signals and specific gesture signals have a unique energy spectrum. Specifically, an improved Linear Discriminant Analysis Algorithm (I-LDA) is devised to reduce the dimension of human behavior signals. Additionally, behaviors are learned by Logistic Regression Algorithm (LRA). Bandwidth ratios in an energy spectrum are selected as features to eliminate the impact of speed differences on results. The system is based on commercial off-the-shelf WiFi devices and we conduct a large number of experiments in a typical indoor environment to evaluate its performance. Experimental results show that our system is robust with average recognition accuracy of up to 96%.

Performance Analysis of Data Mining Algorithms

The globalisation of Internet is creating enormous amount of data on servers. The data created during last two years is itself equivalent to the data created during all these years. This exponential creation of data is due to the easy access to devices based on Internet of things. This information has become a source of predictive analysis for future happenings. The versatile use of computing devices is creating data of diverse nature and the analysts are predicting the future trend using data of their respective domain. The technology used to analyse the data has become a bottleneck over the time. The main reason behind this is that the rate with which the data is getting created is much more than the technology used to access the same. There are various mining techniques used to explore the useful information. In this research there is detailed analysis of how data is used and perceived by various data mining algorithms. Mining algorithms like Naïve Bayes, Support Vector Machines, Linear Discriminant Analysis Algorithm, Artificial Neural Networks, C4.5, C5.0, K-Nearest Neighbour are analysed. The input data used in these algorithms is big data files. This research mainly focuses on how the existing data algorithms are interacting with big data files. The research has been done on twitter comments.

An Efficient Method to Detect Outliers in High Dimensional Data

In this era, detection of outliers or anomalies from high dimensional data is really a great challenge. Normal data is distinguished from data containing anomalies using Outlier detection techniques which classifies new data as normal or abnormal. Different Outlier Detection algorithms are proposed by many researchers for high dimensional data and each algorithm has its own benefits and limitations. In the literature the researchers proposed different algorithms. For this work few algorithms such as Dice-Coefficient Index (DCI), Mapreduce Function and Linear Discriminant Analysis Algorithm (LDA) are considered. Mapreduce function is used to overcome the problem of large datasets. LDA is basically used in the reduction of the data dimensionality. In the present work a novel Hybrid Outlier Detection Algorithm (HbODA) is proposed for efficiently detection of outliers in high dimensional data. The important parameters efficiency, accuracy, computation cost, precision, recall etc. are focused for analyzing the performance of the novel hybrid algorithm. Experimental results on real large sets show that the proposed algorithm is better in detecting outliers than other traditional methods.

Rapid classification of commercial Cheddar cheeses from different brands using PLSDA, LDA and SPA–LDA models built by hyperspectral data

This study aimed to investigate the feasibility of using near-infrared hyperspectral imaging (NIR-HSI) technique for classifying commercial Cheddar cheeses from different brands. Three classification models including a probability based partial least squares discriminant analysis (PLSDA), linear discriminant analysis (LDA) and successive projections algorithm (SPA)–LDA were used to discriminate four brands of Cheddar cheeses. A simple sample ranking method was used to improve the performance of these models. The generalisation abilities of the PLSDA, LDA and SPA–LDA models for the classification of new batch cheeses were investigated. The optimal number of PLS components was 24 for PLSDA, while the optimal SPA selected wavelengths was 105 for SPA–LDA. The results showed that PLSDA built by hyperspectral data was the most suitable model for brands classification with correct classification rate of 86.67%, and SPA–LDA had better performance than LDA with corresponding correct classification rates of 83.33% and 76.67%, respectively. As a comparison, models built by physical data (texture and colour) achieved poor classification results ( < 65% for test set). The current result revealed that HSI technique is better and faster than conventional cheese classification methods.

Microfluidic diffraction phase microscopy for high-throughput, artifact-free quantitative phase imaging and identification of waterborne parasites

An artifact-free microfluidic diffraction phase microscopy (MFDPM) system to measure quantitative phase maps of individual Giardia lamblia (G. lamblia) cysts and Cryptosporidium parvum (C. parvum) oocysts is developed. The system includes a diffraction phase imaging unit aligned with a polydimethylsiloxane (PDMS) microfluidic chip. The microfluidic chip consists of arrays of U-shaped hydrodynamic trapping structures used to trap single cysts/oocysts, and a micromixer to precisely adjust the refractive index of immersion medium in the microfluidic chamber. When the refractive index of immersion medium is tuned to be equal to that of PDMS, the pronounced artifacts in quantitative phase imaging (QPI) originating from diffraction and scattering at the edges of trapping structures can be eliminated. The artifact-free quantitative phase maps of single microparticles are retrieved using Goldstein’s algorithm. Principal component analysis (PCA) is implemented to extract the distinctive features of the quantitative phase maps of G. lamblia cysts and C. parvum oocysts, and they can be identified with 100% accuracy using linear discriminant analysis (LDA) algorithm. The MFDPM is demonstrated to be a simple, accurate and high-throughput technique for artifact-free quantitative phase imaging of waterborne protozoa, and has great potential for label-free and rapid detection and identification of waterborne microorganisms.

Machine Learning Identification of Surgical and Operative Factors Associated With Surgical Expertise in Virtual Reality Simulation

Despite advances in the assessment of technical skills in surgery, a clear understanding of the composites of technical expertise is lacking. Surgical simulation allows for the quantitation of psychomotor skills, generating data sets that can be analyzed using machine learning algorithms. To identify surgical and operative factors selected by a machine learning algorithm to accurately classify participants by level of expertise in a virtual reality surgical procedure. Fifty participants from a single university were recruited between March 1, 2015, and May 31, 2016, to participate in a case series study at McGill University Neurosurgical Simulation and Artificial Intelligence Learning Centre. Data were collected at a single time point and no follow-up data were collected. Individuals were classified a priori as expert (neurosurgery staff), seniors (neurosurgical fellows and senior residents), juniors (neurosurgical junior residents), and medical students, all of whom participated in 250 simulated tumor resections. All individuals participated in a virtual reality neurosurgical tumor resection scenario. Each scenario was repeated 5 times. Through an iterative process, performance metrics associated with instrument movement and force, resection of tissues, and bleeding generated from the raw simulator data output were selected by K-nearest neighbor, naive Bayes, discriminant analysis, and support vector machine algorithms to most accurately determine group membership. A total of 50 individuals (9 women and 41 men; mean [SD] age, 33.6 [9.5] years; 14 neurosurgeons, 4 fellows, 10 senior residents, 10 junior residents, and 12 medical students) participated. Neurosurgeons were in practice between 1 and 25 years, with 9 (64%) involving a predominantly cranial practice. The K-nearest neighbor algorithm had an accuracy of 90% (45 of 50), the naive Bayes algorithm had an accuracy of 84% (42 of 50), the discriminant analysis algorithm had an accuracy of 78% (39 of 50), and the support vector machine algorithm had an accuracy of 76% (38 of 50). The K-nearest neighbor algorithm used 6 performance metrics to classify participants, the naive Bayes algorithm used 9 performance metrics, the discriminant analysis algorithm used 8 performance metrics, and the support vector machine algorithm used 8 performance metrics. Two neurosurgeons, 1 fellow or senior resident, 1 junior resident, and 1 medical student were misclassified. In a virtual reality neurosurgical tumor resection study, a machine learning algorithm successfully classified participants into 4 levels of expertise with 90% accuracy. These findings suggest that algorithms may be capable of classifying surgical expertise with greater granularity and precision than has been previously demonstrated in surgery.

Prediction of Student\u2019s performance by modelling small dataset size

Prediction of student’s performance became an urgent desire in most of educational entities and institutes. That is essential in order to help at-risk students and assure their retention, providing the excellent learning resources and experience, and improving the university’s ranking and reputation. However, that might be difficult to be achieved for startup to mid-sized universities, especially those which are specialized in graduate and post graduate programs, and have small students’ records for analysis. So, the main aim of this project is to prove the possibility of training and modeling a small dataset size and the feasibility of creating a prediction model with credible accuracy rate. This research explores as well the possibility of identifying the key indicators in the small dataset, which will be utilized in creating the prediction model, using visualization and clustering algorithms. Best indicators were fed into multiple machine learning algorithms to evaluate them for the most accurate model. Among the selected algorithms, the results proved the ability of clustering algorithm in identifying key indicators in small datasets. The main outcomes of this study have proved the efficiency of support vector machine and learning discriminant analysis algorithms in training small dataset size and in producing an acceptable classification’s accuracy and reliability test rates.

A Coarse-to-Fine Framework for Cloud Removal in Remote Sensing Image Sequence

Clouds and accompanying shadows, which exist in optical remote sensing images with high possibility, can degrade or even completely occlude certain ground-cover information in images, limiting their applicabilities for Earth observation, change detection, or land-cover classification. In this paper, we aim to deal with cloud contamination problems with the objective of generating cloud-removed remote sensing images. Inspired by low-rank representation together with sparsity constraints, we propose a coarse-to-fine framework for cloud removal in the remote sensing image sequence. Leveraging on group-sparsity constraint, we first decompose the observed cloud image sequence of the same area into the low-rank component, group-sparse outliers, and sparse noise, corresponding to cloud-free land-covers, clouds (and accompanying shadows), and noise respectively. Subsequently, a discriminative robust principal component analysis (RPCA) algorithm is utilized to assign aggressive penalizing weights to the initially detected cloud pixels to facilitate cloud removal and scene restoration. Moreover, we incorporate geometrical transformation into a low-rank model to address the misalignment of the image sequence. Significantly superior to conventional cloud-removal methods, neither cloud-free reference image(s) nor additional operations of cloud and shadow detection are required in our method. Extensive experiments on both simulated data and real data demonstrate that our method works effectively, outperforming many state-of-the-art approaches.

Regularized Group Sparse Discriminant Analysis for P300-Based Brain-Computer Interface.

Event-related potentials (ERPs) especially P300 are popular effective features for brain-computer interface (BCI) systems based on electroencephalography (EEG). Traditional ERP-based BCI systems may perform poorly for small training samples, i.e. the undersampling problem. In this study, the ERP classification problem was investigated, in particular, the ERP classification in the high-dimensional setting with the number of features larger than the number of samples was studied. A flexible group sparse discriminative analysis algorithm based on Moreau-Yosida regularization was proposed for alleviating the undersampling problem. An optimization problem with the group sparse criterion was presented, and the optimal solution was proposed by using the regularized optimal scoring method. During the alternating iteration procedure, the feature selection and classification were performed simultaneously. Two P300-based BCI datasets were used to evaluate our proposed new method and compare it with existing standard methods. The experimental results indicated that the features extracted via our proposed method are efficient and provide an overall better P300 classification accuracy compared with several state-of-the-art methods.

Fréchet mean-based Grassmann discriminant analysis

Representing image sets and videos with Grassmann manifold has become popular due to its powerful capability to extract discriminative information in machine learning research. However, existing techniques operations on Grassmann manifold are usually suffering from the problem of computational expensive, thus the application range of Grassmann manifold is limited. In this paper, we propose the Frechet mean-based Grassmann discriminant analysis (FMGDA) algorithm to implement the videos (or image sets) data dimensionality reduction and clustering task. The data dimensionality reduction algorithm proposed by us can not only be used to reduce Grassmann data from high-dimensional data to a relative low-dimensional data, but also to maximize between-class distance and minimize within-class distance simultaneously. Frechet mean is used to characterize the clustering center of Grassmann manifold space. We further show that the learning problem can be expressed as a trace ratio problem which can be efficiently solved. We designed a detailed experimental scheme to test the performance of our proposed algorithm, and the tests were assessed on several benchmark data sets. The experimental results indicate that our approach leads to a significant improvement over state-of-the-art methods.

A sparse fault degradation oriented fisher discriminant analysis (FDFDA) algorithm for faulty variable isolation and its industrial application

In a fault process, the variables may be influenced differently. In order to improve the diagnosis performance, it is an important issue to isolate those significant faulty variables that cover informative fault effects. However, those variables are selected one by one and their correlations are not considered in the previous work. As sparse-relevant methods can automatically and efficiently isolate significant correlated variables, it is natural to consider applying the criteria of sparsity to separate the significantly influenced faulty variables and analyze them by specific methods. First, the sparse version of the fault degradation oriented Fisher discriminant analysis (FDFDA) algorithm is proposed to produce informative discriminant directions with sparse loadings. Subsequently, a faulty variable selection strategy is proposed based on the sparse FDFDA algorithm to select significantly influenced faulty variables. By iteratively isolating correlated variables along each sparse fault direction, all the faulty variables can be automatically selected until the left fault data and normal data share the similar characteristics. Therefore, the whole measurement variables can be divided into faulty variable set and normal variable set. Then different fault diagnosis models can be developed according to their different characteristics for each fault class. For online application, a probabilistic fault diagnosis strategy is proposed to determine the fault cause of the new sample by the largest synthetic probability that integrates the diagnosis results of two variable sets. The performance of the proposed fault diagnosis method is illustrated using the data from the cut-made process of cigarette.

Discrimination of Tomato Plants (Solanum lycopersicum) Grown under Anaerobic Baffled Reactor Effluent, Nitrified Urine Concentrates and Commercial Hydroponic Fertilizer Regimes Using Simulated Sensor Spectral Settings

We assess the discriminative strength of three different satellite spectral settings (HyspIRI, the forthcoming Landsat 9 and Sentinel 2-MSI), in mapping tomato (Solanum lycopersicum Linnaeus) plants grown under hydroponic system, using human-excreta derived materials (HEDM), namely, anaerobic baffled reactor (ABR) effluent and nitrified urine concentrate (NUC) and commercial hydroponic fertilizer mix (CHFM) as main sources of nutrients. Simulated spectral settings of HyspIRI, Landsat 9 and Sentinel 2-MSI were resampled from spectrometric proximally sensed data. Discriminant analysis (DA) was applied in discriminating tomatoes grown under these different nutrient sources. Results showed that the simulated spectral settings of HyspIRI sensor better discriminate tomatoes grown under different fertilizer regimes when compared to Landsat 9 OLI and Sentinel-2 MSI spectral configurations. Using the DA algorithm, HyspIRI exhibited high overall accuracy (OA) of 0.99 and a kappa statistic of 0.99 whereas Landsat OLI and Sentinel-2 MSI exhibited OA of 0.94 and 0.95 and 0.79 and 0.85 kappa statistics, respectively. Simulated HyspIRI wavebands 710, 720, 690, 840, 1370 and 2110 nm, Sentinel 2-MSI bands 7 (783 nm), 6 (740 nm), 5 (705 nm) and 8a (865 nm) as well as Landsat bands 5 (865 nm), 6 (1610 nm), 7 (2200 nm) and 8 (590 nm), in order of importance, were selected as the most suitable bands for discriminating tomatoes grown under different fertilizer regimes. Overall, the performance of simulated HyspIRI, Landsat 9 OLI-2 and Sentinel-2 MSI spectral bands seem to bring new opportunities for crop monitoring.

An improved Fisher discriminant analysis algorithm based on Procrustes analysis for adaptive fault recognition

Aiming at the problem of continuous model updating for fault recognition in the time-varying process, a novel method called the Procrustes analysis–based Fisher discriminant analysis was proposed. First, each class of the training data was preprocessed by Procrustes analysis. Second, the new test data were aligned with each class of the training data by Procrustes analysis. Then, all the data were reduced to a low-dimensional space using Fisher discriminant analysis. Finally, the Euclidean distance between the test data and the training data after the Procrustes analysis was calculated, and the class recognition was achieved based on the discriminant principle of Fisher discriminant analysis. Two case studies show that the proposed Procrustes analysis–based Fisher discriminant analysis is superior to the traditional method based on Fisher discriminant analysis, and it can be used for fault recognition in a new and efficient way.

Supervised cross-fusion method: a new triplet approach to fuse thermal, radar, and optical satellite data for land use classification

This study presents a new fusion method namely supervised cross-fusion method to improve the capability of fused thermal, radar, and optical images for classification. The proposed cross-fusion method is a combination of pixel-based and supervised feature-based fusion of thermal, radar, and optical data. The pixel-based fusion was applied to fuse optical data of Sentinel-2 and Landsat 8. According to correlation coefficient (CR) and signal to noise ratio (SNR), among the used pixel-based fusion methods, wavelet obtained the best results for fusion. Considering spectral and spatial information preservation, CR of the wavelet method is 0.97 and 0.96, respectively. The supervised feature-based fusion method is a fusion of best output of pixel-based fusion level, land surface temperature (LST) data, and Sentinel-1 radar image using a supervised approach. The supervised approach is a supervised feature selection and learning of the inputs based on linear discriminant analysis and sparse regularization (LDASR) algorithm. In the present study, the non-negative matrix factorization (NMF) was utilized for feature extraction. A comparison of the obtained results with state of the art fusion method indicated a higher accuracy of our proposed method of classification. The rotation forest (RoF) classification results improvement was 25% and the support vector machine (SVM) results improvement was 31%. The results showed that the proposed method is well classified and separated four main classes of settlements, barren land, river, river bank, and even the bridges over the river. Also, a number of unclassified pixels by SVM are very low compared to other classification methods and can be neglected. The study results showed that LST calculated using thermal data has had positive effects on improving the classification results. By comparing the results of supervised cross-fusion without using LST data to the proposed method results, SVM and RoF classifiers showed 38% and 7% of classification improvement, respectively.

Assessment of External Properties for Identifying Banana Fruit Maturity Stages Using Optical Imaging Techniques

The maturity stage of bananas has a considerable influence on the fruit postharvest quality and the shelf life. In this study, an optical imaging based method was formulated to assess the importance of different external properties on the identification of four successive banana maturity stages. External optical properties, including the peel color and the local textural and local shape information, were extracted from the stalk, middle and tip of the bananas. Specifically, the peel color attributes were calculated from individual channels in the hue-saturation-value (HSV), the International Commission on Illumination (CIE) L*a*b* and the CIE L*ch color spaces; the textural information was encoded using a local binary pattern with uniform patterns (UP-LBP); and the local shape features were described by histogram of oriented gradients (HOG). Three classifiers based on the naïve Bayes (NB), linear discriminant analysis (LDA) and support vector machine (SVM) algorithms were adopted to evaluate the performance of identifying banana fruit maturity stages using the different optical appearance features. The experimental results demonstrate that overall identification accuracies of 99.2%, 100% and 99.2% were achieved using color appearance features with the NB, LDA and SVM classifiers, respectively; overall accuracies of 92.6%, 86.8% and 93.4% were obtained using local textural features for the three classifiers, respectively; and overall accuracies of only 84.3%, 83.5% and 82.6% were obtained using local shape features with the three classifiers, respectively. Compared to the complicated calculation of both the local textural and local shape properties, the simplicity and high accuracy of the peel color property make it more appropriate for identifying banana fruit maturity stages using optical imaging techniques.