Search results for "Machine Learning"

Reduction of the number of spectral bands in Landsat images: a comparison of linear and nonlinear methods

2006

We describe some applications of linear and nonlinear pro- jection methods in order to reduce the number of spectral bands in Land- sat multispectral images. The nonlinear method is curvilinear component analysis CCA, and we propose an adapted optimization of it for image processing, based on the use of principal-component analysis PCA, a linear method. The principle of CCA consists in reproducing the topol- ogy of the original space projection points in a reduced subspace, keep- ing the maximum of information. Our conclusions are: CCA is an im- provement for dimension reduction of multispectral images; CCA is really a nonlinear extension of PCA; CCA optimization through PCA called CCAinitP…

Comparing ELM Against MLP for Electrical Power Prediction in Buildings

2015

The study of energy efficiency in buildings is an active field of research. Modelling and predicting energy related magnitudes leads to analyse electric power consumption and can achieve economical benefits. In this study, two machine learning techniques are applied to predict active power in buildings. The real data acquired corresponds to time, environmental and electrical data of 30 buildings belonging to the University of Leon (Spain). Firstly, we segmented buildings in terms of their energy consumption using principal component analysis. Afterwards we applied ELM and MLP methods to compare their performance. Models were studied for different variable selections. Our analysis shows that…

Machine Learning Approaches for Environmental Mixtures Studies with Time-to-Event Outcomes and their Application to the Strong Heart Study

2021

CrowdVAS-Net: A Deep-CNN Based Framework to Detect Abnormal Crowd-Motion Behavior in Videos for Predicting Crowd Disaster

2019

With the increased occurrences of crowd disasters like human stampedes, crowd management and their safety during mass gathering events like concerts, congregation or political rally, etc., are vital tasks for the security personnel. In this paper, we propose a framework named as CrowdVAS-Net for crowd-motion analysis that considers velocity, acceleration and saliency features in the video frames of a moving crowd. CrowdVAS-Net relies on a deep convolutional neural network (DCNN) for extracting motion and appearance feature representations from the video frames that help us in classifying the crowd-motion behavior as abnormal or normal from a short video clip. These feature representations a…

Comprehensive Experimental Analysis of Handcrafted Descriptors for Face Recognition

2018

Over the past few decades, LBP descriptor, which shown its high robustness in extracting discriminative features from an image, has been successfully applied in diverse challenging computer vision applications including face recognition. The efficiency and usability of the LBP operator and its success in various real world applications has inspired the development of much new powerful LBP variants. Indeed, after the appearance of the LBP operator, several renowned extensions and modifications of LBP have been proposed in the literature to the point that it can be difficult to recognize their respective LBP-related strategies, strengths and weaknesses according to a given application, and th…

Automatic place detection and localization in autonomous robotics

2007

This paper presents an approach for the simultaneous learning and recognition of places applied to autonomous robotics. While noteworthy results have been achieved with respect to off-line training process for appearance-based navigation, novel issues arise when recognition and learning are simultaneous and unsupervised processes. The approach adopted here uses a Gaussian mixture model estimated by a novel incremental MML-EM to model the probability distribution of features extracted by image-preprocessing. A place detector decides which features belong to which place integrating odometric information and a hidden Markov model. Tests demonstrate that the proposed system performs as well as …

Classification Similarity Learning Using Feature-Based and Distance-Based Representations: A Comparative Study

2015

Automatically measuring the similarity between a pair of objects is a common and important task in the machine learning and pattern recognition fields. Being an object of study for decades, it has lately received an increasing interest from the scientific community. Usually, the proposed solutions have used either a feature-based or a distance-based representation to perform learning and classification tasks. This article presents the results of a comparative experimental study between these two approaches for computing similarity scores using a classification-based method. In particular, we use the Support Vector Machine as a flexible combiner both for a high dimensional feature space and …

An improved distance-based relevance feedback strategy for image retrieval

2013

Most CBIR (content based image retrieval) systems use relevance feedback as a mechanism to improve retrieval results. NN (nearest neighbor) approaches provide an efficient method to compute relevance scores, by using estimated densities of relevant and non-relevant samples in a particular feature space. In this paper, particularities of the CBIR problem are exploited to propose an improved relevance feedback algorithm based on the NN approach. The resulting method has been tested in a number of different situations and compared to the standard NN approach and other existing relevance feedback mechanisms. Experimental results evidence significant improvements in most cases.

Evaluating State-Based Intention Recognition Algorithms against Human Performance

2014

In this paper, we describe a novel intention recognition approach based on the representation of state information in a cooperative human-robot environment. We compare the output of the intention recognition algorithms to those of an experiment involving humans attempting to recognize the same intentions in a manufacturing kitting domain. States are represented by a combination of spatial relationships in a Cartesian frame along with cardinal direction information. Based upon a set of predefined high-level states relationships that must be true for future actions to occur, a robot can use the approaches described in this paper to infer the likelihood of subsequent actions occurring. This wo…

Classification Boundary Approximation by Using Combination of Training Steps for Real-Time Image Segmentation

2007

We propose a method of real-time implementation of an approximation of the support vector machine decision rule. The method uses an improvement of a supervised classification method based on hyperrectangles, which is useful for real-time image segmentation. We increase the classification and speed performances using a combination of classification methods: a support vector machine is used during a pre-processing step. We recall the principles of the classification methods and we evaluate the hardware implementation cost of each method. We present our learning step combination algorithm and results obtained using Gaussian distributions and an example of image segmentation coming from a part …