Representative Training Samples Research Articles

Construction of Membership Functions for Soil Mapping using the Partial Dependence of Soil on Environmental Covariates Calculated by Random Forest

Core Ideas This study develops a method to construct membership functions representing knowledge on soil–environment relationships from partial dependence. Use of representative samples as training samples is recommended when applying the proposed method. Training samples (including representative samples and other samples) with good coverage in the environmental feature space would allow Random Forest to obtain more accurate soil maps than using representative samples. Partial dependence plots generated by Random Forest imply an association between soil and environmental variables. Partial dependence plots generated by Random Forest (RF) imply an association between soil and environmental variables. This study develops a method to construct membership functions representing knowledge of soil–environment relationships from partial dependence. Key parameters were obtained from normalized partial dependence to define class limits and membership gradation. Seven environmental variables were selected on the basis of the variable's importance within RF. Two cases were conducted to test the effectiveness of our method using different training samples. Case 1 used 33 representative locations as training samples and 50 locations as validations. Case 2 randomly split all 83 samples into training and validation subsets at a proportion of 2:1; the splits were repeated seven times. For each case, the generated membership functions were used for mapping soil subgroups in Heshan, China, under the Soil Landscape Inference Model framework; RF was conducted for comparison. The results showed that mapping accuracy based on the membership functions (78%) was much higher than that of RF only (60%) in Case 1. In Case 2, the mapping accuracies using membership functions (an average of 67%, SD = 6.5%) were not always higher than those by RF (an average of 67%, SD = 8.0%). The constructed membership functions were impacted by the training samples. Use of representative training samples is recommended when applying the proposed method. However, training samples (including representative samples and other samples) with good coverage in the environmental feature space would allow RF to obtain more accurate soil maps than using representative samples.

Martial Arts, Dancing and Sports dataset: A challenging stereo and multi-view dataset for 3D human pose estimation

Human pose estimation is one of the most popular research topics in the past two decades, especially with the introduction of human pose datasets for benchmark evaluation. These datasets usually capture simple daily life actions. Here, we introduce a new dataset, the Martial Arts, Dancing and Sports (MADS), which consists of challenging martial arts actions (Tai-chi and Karate), dancing actions (hip-hop and jazz), and sports actions (basketball, volleyball, football, rugby, tennis and badminton). Two martial art masters, two dancers and an athlete performed these actions while being recorded with either multiple cameras or a stereo depth camera. In the multi-view or single-view setting, we provide three color views for 2D image-based human pose estimation algorithms. For depth-based human pose estimation, we provide stereo-based depth images from a single view. All videos have corresponding synchronized and calibrated ground-truth poses, which were captured using a Motion Capture system. We provide initial baseline results on our dataset using a variety of tracking frameworks, including a generative tracker based on the annealing particle filter and robust likelihood function, a discriminative tracker using twin Gaussian processes [1], and hybrid trackers, such as Personalized Depth Tracker [2]. The results of our evaluation suggest that discriminative approaches perform better than generative approaches when there are enough representative training samples, and that the generative methods are more robust to diversity of poses, but can fail to track when the motion is too quick for the effective search range of the particle filter. The data and the accompanying code will be made available to the research community.

Filter selection based on representative training samples for multispectral imaging

In the application of spectral reflectance estimation with multispectral imaging, the filters and training samples jointly influence the accuracy of spectral estimation for learning based methods. In this study, by utilizing one kind of genetic algorithm, the selection of optimal filters is implemented with respect to the representative training samples so as to consider the spectral characteristics of filters, imaging system and test samples simultaneously, which also avoids measurement on the detailed spectral characteristics of filters and imaging system. This filter selection method was tested on a filter-wheel based multispectral imaging device along with a set of interference filters. It is verified that a small number of representative training samples are sufficient for selecting the optimal filters, and the filters selected by this method are more effective for spectral estimation than those obtained from the conventional method.

A combination of chemometrics methods and GC–MS for the classification of edible vegetable oils

The authenticity of edible vegetable oils is a very important issue due to consumer health and commercial reasons. Gas chromatography–mass spectrometry (GC–MS) was applied to analyze the fatty acid composition of sixty six samples from six different kinds of edible vegetable oils. The fatty acid profiles of these edible vegetable oils were used to classify the type of edible oils. For improving the classification accuracy of vegetable oils with respect to type, the support vector machine (SVM) technique, optimized using the genetic algorithm (GA), was employed to construct the classification model. The effectiveness of the GA–SVM combination in classification was compared with that of other well-known strategies for classification, such as minimum distance classification (MDC) and linear discriminant analysis (LDA). In addition, the Kennard–Stone algorithm was used to select the representative training samples and compared with the random sampling method. The misclassification rates were 8.48% and 3.03% for training and test set, respectively, by the GA–SVM model using the linear kernel. Only one or two samples will be misclassified in the process of GA–SVM classification. The classification task based on fatty acid data can be successfully achieved by the GA–SVM technique combined with the Kennard–Stone algorithm. The results reveal that this strategy is of great promise in flexible and accurate classification of edible vegetable oils.

A new research paradigm for global land cover mapping

ABSTRACTIn this paper, we introduced major challenges in mapping croplands, settlements, water and wetlands, and discussed challenges in the use of multi-temporal and multi-sensor data. We then summarized some of the on-going efforts in improving qualities of global land cover maps. Existing technologies provide sufficient data for better map making if extra efforts can be made instead of harmonizing and integrating various global land cover products. Developing and selecting better algorithms, including more input variables (new types of data or features) for classification, having representative training samples are among conventional measures generally believed effective in improving mapping accuracies at local scales. We pointed out that data were more important in improving mapping accuracies than algorithms. Finally, we proposed a new paradigm for global land cover mapping, which included a view of vegetation classes based on their types and form, canopy cover and height. The new paradigm suggests that a universally applicable training sample set is not only possible but also effective in improving land cover classification at the continental and global scales. To ensure an easy transition from traditional land cover mapping to the new paradigm, we recommended that an all-in-one data management and analysis system be constructed.

Enhanced Local Fisher Discriminant Analysis for Indoor Positioning in Wireless Local Area Network

Feature extraction methods have been used to extract location features for indoor positioning in wireless local area networks. However, existing methods, such as linear discriminant analysis and principal component analysis, all suffer from the multimodal property of signal distribution. This paper proposes a novel method, based on enhanced local fisher discriminant analysis (LFDA). First, LFDA is proposed to extract discriminative location features. It maximizes between-class separability while preserving within-class local structure of signal space, thereby guaranteeing maximal discriminative information involved in positioning. Then, the generalization ability of LFDA is further enhanced using signal perturbation, which generates more number of representative training samples. Experimental results in realistic indoor environment show that, compared with previous feature extraction methods, the proposed method reduces the mean and standard deviation of positing error by 23.9% and 33.0%, respectively.

Sparse and Dense Hybrid Representation via Dictionary Decomposition for Face Recognition.

Sparse representation provides an effective tool for classification under the conditions that every class has sufficient representative training samples and the training data are uncorrupted. These conditions may not hold true in many practical applications. Face identification is an example where we have a large number of identities but sufficient representative and uncorrupted training images cannot be guaranteed for every identity. A violation of the two conditions leads to a poor performance of the sparse representation-based classification (SRC). This paper addresses this critic issue by analyzing the merits and limitations of SRC. A sparse- and dense-hybrid representation (SDR) framework is proposed in this paper to alleviate the problems of SRC. We further propose a procedure of supervised low-rank (SLR) dictionary decomposition to facilitate the proposed SDR framework. In addition, the problem of the corrupted training data is also alleviated by the proposed SLR dictionary decomposition. The application of the proposed SDR-SLR approach in face recognition verifies its effectiveness and advancement to the field. Extensive experiments on benchmark face databases demonstrate that it consistently outperforms the state-of-the-art sparse representation based approaches and the performance gains are significant in most cases.

A comparative analysis of approaches for successional vegetation classification in the Brazilian Amazon

Research on separation of successional stages has been an active topic for the past two decades because successional vegetation plays an important role in the carbon budget and restoration of soil fertility in the Brazilian Amazon. This article examines classification of successional stages by conducting a comparative analysis of classification algorithms (maximum likelihood classifier – MLC, artificial neural network – ANN, K-nearest neighbour – KNN, support vector machine – SVM, classification tree analysis – CTA, and object-based classification – OBC) on varying remote-sensing data-sets (Landsat and ALOS PALSAR). Through this research we obtained the following four major conclusions: (1) Landsat data provide higher classification accuracy than ALOS PALSAR data, and individual PALSAR data cannot effectively separate successional stages; (2) Fusion of Landsat and PALSAR data provides better classification than individual sensor data; (3) Depending on the data-set, the best classification algorithm varies, MLC and CTA are recommended for Landsat or fusion images; and KNN is recommended for the combination of Landsat and PALSAR data as extra bands; (4) the MLC based on fusion images is recommended for vegetation classification in the moist tropical region when sufficiently representative training samples are available.

Automatic Myonuclear Detection in Isolated Single Muscle Fibers Using Robust Ellipse Fitting and Sparse Representation.

Accurate and robust detection of myonuclei in isolated single muscle fibers is required to calculate myonuclear domain size. However, this task is challenging because: 1) shape and size variations of the nuclei, 2) overlapping nuclear clumps, and 3) multiple z-stack images with out-of-focus regions. In this paper, we have proposed a novel automatic detection algorithm to robustly quantify myonuclei in isolated single skeletal muscle fibers. The original z-stack images are first converted into one all-in-focus image using multi-focus image fusion. A sufficient number of ellipse fitting hypotheses are then generated from the myonuclei contour segments using heteroscedastic errors-in-variables (HEIV) regression. A set of representative training samples and a set of discriminative features are selected by a two-stage sparse model. The selected samples with representative features are utilized to train a classifier to select the best candidates. A modified inner geodesic distance based mean-shift clustering algorithm is used to produce the final nuclei detection results. The proposed method was extensively tested using 42 sets of z-stack images containing over 1,500 myonuclei. The method demonstrates excellent results that are better than current state-of-the-art approaches.

Automatic Ki-67 Counting Using Robust Cell Detection and Online Dictionary Learning

Ki-67 proliferation index is a valid and important biomarker to gauge neuroendocrine tumor (NET) cell progression within the gastrointestinal tract and pancreas. Automatic Ki-67 assessment is very challenging due to complex variations of cell characteristics. In this paper, we propose an integrated learning-based framework for accurate automatic Ki-67 counting for NET. The main contributions of our method are: 1) A robust cell counting and boundary delineation algorithm that is designed to localize both tumor and nontumor cells. 2) A novel online sparse dictionary learning method to select a set of representative training samples. 3) An automated framework that is used to differentiate tumor from nontumor cells (such as lymphocytes) and immunopositive from immunonegative tumor cells for the assessment of Ki-67 proliferation index. The proposed method has been extensively tested using 46 NET cases. The performance is compared with pathologists' manual annotations. The automatic Ki-67 counting is quite accurate compared with pathologists' manual annotations. This is much more accurate than existing methods.

Comparison of Classification Algorithms and Training Sample Sizes in Urban Land Classification with Landsat Thematic Mapper Imagery

Although a large number of new image classification algorithms have been developed, they are rarely tested with the same classification task. In this research, with the same Landsat Thematic Mapper (TM) data set and the same classification scheme over Guangzhou City, China, we tested two unsupervised and 13 supervised classification algorithms, including a number of machine learning algorithms that became popular in remote sensing during the past 20 years. Our analysis focused primarily on the spectral information provided by the TM data. We assessed all algorithms in a per-pixel classification decision experiment and all supervised algorithms in a segment-based experiment. We found that when sufficiently representative training samples were used, most algorithms performed reasonably well. Lack of training samples led to greater classification accuracy discrepancies than classification algorithms themselves. Some algorithms were more tolerable to insufficient (less representative) training samples than others. Many algorithms improved the overall accuracy marginally with per-segment decision making.

A new sparse representation-based classification algorithm using iterative class elimination

In this paper, we propose a sparse representation-based classification method using iterative class elimination strategy for face recognition. The proposed method aims to represent a test sample as a linear combination of the most competitive training samples and exploits an optimal representation of training samples from the classes with major relevant contributions. We interpret the sparse representation problem as an information fidelity problem. In the context of our proposed method, an important goal is to select a subset of variables for accomplishing one objective: the provision of a descriptive representation for sparse class knowledge structure. We develop an iterative class elimination algorithm to achieve this goal. First, the contribution in presenting the test sample of any of the specified classes is, respectively, calculated by adding up the total contribution of all training samples of this class, and then a certain class that meets the smallest score requirement to this test sample is eliminated. Second, a similar procedure is iteratively carried out for the set of remaining training samples from rest classes, and this procedure is repeatedly performed till the predefined termination condition is satisfied. The final remaining training samples are used to produce a best representation of the test sample and to classify it. Therefore, the proposed algorithm is an iterative method that alternates between sparse representation and a process of updating the training atoms to better fit the test data. This is helpful to accurately classify the test sample. Experimental results conducted on the ORL, FERET, and AR face databases demonstrate the effectiveness of the proposed method.

Identifying predictive hubs to condense the training set of $$k$$ -nearest neighbour classifiers

The $$k$$ -Nearest Neighbour classifier is widely used and popular due to its inherent simplicity and the avoidance of model assumptions. Although the approach has been shown to yield a near-optimal classification performance for an infinite number of samples, a selection of the most decisive data points can improve the classification accuracy considerably in real settings with a limited number of samples. At the same time, a selection of a subset of representative training samples reduces the required amount of storage and computational resources. We devised a new approach that selects a representative training subset on the basis of an evolutionary optimization procedure. This method chooses those training samples that have a strong influence on the correct prediction of other training samples, in particular those that have uncertain labels. The performance of the algorithm is evaluated on different data sets. Additionally, we provide graphical examples of the selection procedure.

Appearance-based representative samples refining method for palmprint recognition

The sparse representation can deal with the lack of sample problem due to utilizing of all the training samples. However, the discrimination ability will degrade when more training samples are used for representation. We propose a novel appearance-based palmprint recognition method. We aim to find a compromise between the discrimination ability and the lack of sample problem so as to obtain a proper representation scheme. Under the assumption that the test sample can be well represented by a linear combination of a certain number of training samples, we first select the representative training samples according to the contributions of the samples. Then we further refine the training samples by an iteration procedure, excluding the training sample with the least contribution to the test sample for each time. Experiments on PolyU multispectral palmprint database and two-dimensional and three-dimensional palmprint database show that the proposed method outperforms the conventional appearance-based palmprint recognition methods. Moreover, we also explore and find out the principle of the usage for the key parameters in the proposed algorithm, which facilitates to obtain high-recognition accuracy.

Learning-Based Prediction of Visual Attention for Video Signals

Visual attention, which is an important characteristic of human visual system, is a useful clue for image processing and compression applications in the real world. This paper proposes a computational scheme that adopts both low-level and high-level features to predict visual attention from video signal by machine learning. The adoption of low-level features (color, orientation, and motion) is based on the study of visual cells, and the adoption of the human face as a high-level feature is based on the study of media communications. We show that such a scheme is more robust than those using purely single low- or high-level features. Unlike conventional techniques, our scheme is able to learn the relationship between features and visual attention to avoid perceptual mismatch between the estimated salience and the actual human fixation. We also show that selecting the representative training samples according to the fixation distribution improves the efficacy of regressive training. Experimental results are shown to demonstrate the advantages of the proposed scheme.

Statistical classification techniques for photometric supernova typing

Future photometric supernova surveys will produce vastly more candidates than can be followed up spectroscopically, highlighting the need for effective classification methods based on lightcurves alone. Here we introduce boosting and kernel density estimation techniques which have minimal astrophysical input, and compare their performance on 20,000 simulated Dark Energy Survey lightcurves. We demonstrate that these methods are comparable to the best template fitting methods currently used, and in particular do not require the redshift of the host galaxy or candidate. However both methods require a training sample that is representative of the full population, so typical spectroscopic supernova subsamples will lead to poor performance. To enable the full potential of such blind methods, we recommend that representative training samples should be used and so specific attention should be given to their creation in the design phase of future photometric surveys.

Automatic configuration of optimized sample-weighted least-squares support vector machine by particle swarm optimization for multivariate spectral analysis

Due to the high dimensionality and complexity of multivariate spectral data space and the uncertainty involved in the sampling process, the representation of training samples in the whole sample space is difficult to evaluate and selection of representative training samples for conventional multivariate calibration depends largely on experiential methods. If the training samples fail to represent the sample space, sometimes the prediction of new samples can be degraded. To circumvent this problem, in this paper, a new optimized sample-weighted least-squares support vector machine (OSWLS-SVM) multivariate calibration method is proposed by incorporating the concept of weighted sampling into LS-SVM, where the complexity and predictivity of the model are considered simultaneously. A recently suggested global optimization technique base on particle swarm optimization (PSO) is invoked to simultaneously search for the best sample weights and the hyper-parameters involved in OSWLS-SVM optimizing the training of a calibration set and the prediction of an independent validation set. The implementation of PSO achieves complete automatization of the OSWLS-SVM modeling process and high efficiency in convergence to a desired optimum. Three real multivariate spectral data sets including two public data sets and an experimental data set are investigated and the results are compared favorably with those of PLS and LS-SVM to demonstrate the advantages of the proposed method. The stability and efficiency of OSWLS-SVM is also surveyed, the results reveal that the proposed method can obtain desirable results within moderate PSO cycles.

Incremental learning of feature space and classifier for face recognition

We have proposed a new approach to pattern recognition in which not only a classifier but also a feature space of input variables is learned incrementally. In this paper, an extended version of Incremental Principal Component Analysis (IPCA) and Resource Allocating Network with Long-Term Memory (RAN-LTM) are effectively combined to implement this idea. Since IPCA updates a feature space incrementally by rotating the eigen-axes and increasing the dimensions, the inputs of a neural classifier must also change in their values and the number of input variables. To solve this problem, we derive an approximation of the update formula for memory items, which correspond to representative training samples stored in the long-term memory of RAN-LTM. With these memory items, RAN-LTM is efficiently reconstructed and retrained to adapt to the evolution of the feature space. This function is incorporated into our face recognition system. In the experiments, the proposed incremental learning model is evaluated over a self-compiled video clip of 24 persons. The experimental results show that the incremental learning of a feature space is very effective to enhance the generalization performance of a neural classifier in a realistic face recognition task.

Component-based subspace linear discriminant analysis method for face recognition with one training sample

Many face recognition algorithms/systems have been developed in the last decade and excellent performances have also been reported when there is a sufficient number of representative training samples. In many real-life applications such as passport identification, only one well-controlled frontal sample image is available for training. Under this situation, the performance of existing algorithms will degrade dramatically or may not even be implemented. We propose a component-based linear discriminant analysis (LDA) method to solve the one training sample problem. The basic idea of the proposed method is to construct local facial feature component bunches by moving each local feature region in four directions. In this way, we not only generate more samples with lower dimension than the original image, but also consider the face detection localization error while training. After that, we propose a subspace LDA method, which is tailor-made for a small number of training samples, for the local feature projection to maximize the discrimination power. Theoretical analysis and experiment results show that our proposed subspace LDA is efficient and overcomes the limitations in existing LDA methods. Finally, we combine the contributions of each local component bunch with a weighted combination scheme to draw the recognition decision. A FERET database is used for evaluating the proposed method and results are encouraging.

Mapping soil landscape as spatial continua: The Neural Network Approach

A neural network approach was developed to populate a soil similarity model that was designed to represent soil landscape as spatial continua for hydroecological modeling at watersheds of mesoscale size. The approach employs multilayer feed forward neural networks. The input to the network was data on a set of soil formative environmental factors; the output from the network was a set of similarity values to a set of prescribed soil classes. The network was trained using a conjugate gradient algorithm in combination with a simulated annealing technique to learn the relationships between a set of prescribed soils and their environmental factors. Once trained, the network was used to compute for every location in an area the similarity values of the soil to the set of prescribed soil classes. The similarity values were then used to produce detailed soil spatial information. The approach also included a Geographic Information System procedure for selecting representative training and testing samples and a process of determining the network internal structure. The approach was applied to soil mapping in a watershed, the Lubrecht Experimental Forest, in western Montana. The case study showed that the soil spatial information derived using the neural network approach reveals much greater spatial detail and has a higher quality than that derived from the conventional soil map. Implications of this detailed soil spatial information for hydroecological modeling at the watershed scale are also discussed.