Fábio Augusto Faria

Learning to rank for content-based image retrieval

In Content-based Image Retrieval (CBIR), accurately ranking the returned images is of paramount importance, since users consider mostly the topmost results. The typical ranking strategy used by many CBIR systems is to employ image content descriptors, so that returned images that are most similar to the query image are placed higher in the rank. While this strategy is well accepted and widely used, improved results may be obtained by combining multiple image descriptors. In this paper we explore this idea, and introduce algorithms that learn to combine information coming from different descriptors. The proposed learning to rank algorithms are based on three diverse learning techniques: Support Vector Machines (CBIR-SVM), Genetic Programming (CBIR-GP), and Association Rules (CBIR-AR). Eighteen image content descriptors(color, texture, and shape information) are used as input and provided as training to the learning algorithms. We performed a systematic evaluation involving two complex and heterogeneous image databases (Corel e Caltech) and two evaluation measures (Precision and MAP). The empirical results show that all learning algorithms provide significant gains when compared to the typical ranking strategy in which descriptors are used in isolation. We concluded that, in general, CBIR-AR and CBIR-GP outperforms CBIR-SVM. A fine-grained analysis revealed the lack of correlation between the results provided by CBIR-AR and the results provided by the other two algorithms, which indicates the opportunity of an advantageous hybrid approach.

Automatic identification of fruit flies (Diptera

Description and learning techniques for fruit fly identification.Image analysis techniques for classifying wings and aculeus.A novel multimodal classification approach for fruit fly identification based on wing and aculeus.Good results in the identification of fruit flies based on wings, aculei, and their combination. Fruit flies are pests of major economic importance in agriculture. Among these pests it is possible to highlight some species of genus Anastrepha, which attack a wide range of fruits, and are widely distributed in the American tropics and subtropics. Researchers seek to identify fruit flies in order to implement management and control programs as well as quarantine restrictions. However, fruit fly identification is manually performed by scarce specialists through analysis of morphological features of the mesonotum, wing, and aculeus. Our objective is to find solid knowledge that can serve as a basis for the development of a sounding automatic identification system of the Anastrepha fraterculus group, which is of high economic importance in Brazil. Wing and aculeus images datasets from three specimens have been used in this work. The experiments using a classifier multimodal fusion approach shows promising effectiveness results for identification of these fruit flies, with more than 98% classification accuracy, a remarkable result for this difficult problem.

Illuminant-Based Transformed Spaces for Image Forensics

In this paper, we explore transformed spaces, represented by image illuminant maps, to propose a methodology for selecting complementary forms of characterizing visual properties for an effective and automated detection of image forgeries. We combine statistical telltales provided by different image descriptors that explore color, shape, and texture features. We focus on detecting image forgeries containing people and present a method for locating the forgery, specifically, the face of a person in an image. Experiments performed on three different open-access data sets show the potential of the proposed method for pinpointing image forgeries containing people. In the two first data sets (DSO-1 and DSI-1), the proposed method achieved a classification accuracy of 94% and 84%, respectively, a remarkable improvement when compared with the state-of-the-art methods. Finally, when evaluating the third data set comprising questioned images downloaded from the Internet, we also present a detailed analysis of target images.

A Genetic Programming approach for coffee crop recognition

This work presents a new approach for automatic recognition of coffee crops in RSIs. The method applies an approach based on Genetic Programming (GP) to combine texture and spectral information encoded by image descriptors. Experiments show that the proposed method yields slightly better results than the traditional MaxVer approach.

Automatic Classifier Fusion for Produce Recognition

Recognizing different kinds of fruits and vegetables is a common task in supermarkets. This task, however, poses several challenges as it requires the identification of different species of a particular produce and also its variety. Usually, existing computer-based recognition approaches are not automatic and demand long-term and laborious prior training sessions. This paper presents a novel framework for classifier fusion aiming at supporting the automatic recognition of fruits and vegetables in a supermarket environment. The objective is to provide an effective mechanism for combining low-cost classifiers trained for specific classes of interest. The experiments performed demonstrate that the proposed framework yields better results than several related work found in the literature and represents a step forward automatic produce recognition in cashiers of supermarkets.

Fusion of time series representations for plant recognition in phenology studies

Abstract Nowadays, global warming and its resulting environmental changes is a hot topic in different biology research area. Phenology is one effective way of tracking such environmental changes through the study of plant’s periodic events and their relationship to climate. One promising research direction in this area relies on the use of vegetation images to track phenology changes over time. In this scenario, the creation of effective image-based plant identification systems is of paramount importance. In this paper, we propose the use of a new representation of time series to improve plants recognition rates. This representation, called recurrence plot (RP), is a technique for nonlinear data analysis, which represents repeated events on time series into a two-dimensional representation (an image). Therefore, image descriptors can be used to characterize visual properties from this RP images so that these features can be used as input of a classifier. To the best of our knowledge, this is the first work that uses recurrence plot for plant recognition task. Performed experiments show that RP can be a good solution to describe time series. In addition, in a comparison with visual rhythms (VR), another technique used for time series representation, RP shows a better performance to describe texture properties than VR. On the other hand, a correlation analysis and the adoption of a well successful classifier fusion framework show that both representations provide complementary information that is useful for improving classification accuracies.

Rank Aggregation for Pattern Classifier Selection in Remote Sensing Images

In the past few years, segmentation and classification techniques have become a cornerstone of many successful remote sensing algorithms aiming at delineating geographic target objects. One common strategy relies on using multiple complex features to guide the delineation process with the objective of gathering complementary information for improving classification results. However, a persistent problem in this approach is how to combine different and noncorrelated feature descriptors automatically. In this regard, one solution is to combine them through multiple classifier systems (MCSs) in which the diversity of simple/noncomplex classifiers is an essential issue in the definition of appropriate strategies for classifier fusion. In this paper, we propose a novel strategy for selecting classifiers (whereby a classifier is taken as a pair of learning method plus image descriptor) to be combined in MCS. In the proposed solution, diversity measures are used to assess the degree of agreement/disagreement between pairs of classifiers and ranked lists are created to sort them according to their diversity score. Thereafter, the classifiers are also sorted according to their performance through different evaluation measures (e.g., kappa and tau indices). In the end, a rank aggregation method is proposed to select the most suitable classifiers based on both the diversity and the effectiveness performance of classifiers. The proposed fusion framework has targeted at coffee crop classification and urban recognition but it is general enough to be used in a variety of other pattern recognition problems. Experimental results demonstrate that the novel strategy yields good results when compared to several baselines while using fewer classifiers and being much more efficient.

Improving Image Classification through Descriptor Combination

The efficiency in image classification tasks can be improved using combined information provided by several sources, such as shape, color, and texture visual properties. Although many works proposed to combine different feature vectors, we model the descriptor combination as an optimization problem to be addressed by evolutionary-based techniques, which compute distances between samples that maximize their separability in the feature space. The robustness of the proposed technique is assessed by the Optimum-Path Forest classifier. Experiments showed that the proposed methodology can outperform individual information provided by single descriptors in well-known public datasets.

Classifier Selection Based on the Correlation of Diversity Measures: When Fewer Is More

The ever-growing access to high-resolution images has prompted the development of region-based classification methods for remote sensing images. However, in agricultural applications, the recognition of specific regions is still a challenge as there could be many different spectral patterns in a same studied area. In this context, depending on the features used, different learning methods can be used to create complementary classifiers. Many researchers have developed solutions based on the use of machine learning techniques to address these problems. Examples of successful initiatives are those dedicated to the development of learning techniques for data fusion or Multiple Classifier Systems (MCS). In MCS, diversity becomes an essential factor for their success. Different works have been using diversity measures to select appropriate high-performance classifiers, but the challenge of finding the optimal number of classifiers for a target task has not been properly addressed yet. In general, the proposed solutions rely on the a priori use of ad hoc strategies for selecting classifiers, followed by the evaluation of their effectiveness results during training. Searching by the optimal number of classifiers, however, makes the selection process more expensive. In this paper, we address this issue by proposing a novel strategy for selecting classifiers to be combined based on the correlation of different diversity measures. Diversity measures are used to rank pairs of classifiers and the agreement among ranked lists guides the classifier selection process. A fusion framework has been used in our experiments targeted to the classification of coffee crops in remote sensing images. Experiment results demonstrate that the novel strategy is able to yield comparable effectiveness results when contrasted to several baselines, but using much fewer classifiers.

Automatic fusion of region-based classifiers for coffee crop recognition

Coffee crop recognition in remote sensing images is a complex task. It poses several challenges due to different spectral responses and texture patterns that can be extracted from coffee regions. This paper presents a novel framework for combining different classifiers using support vector machine technique (SVM), which try to learn with each one of classifiers previews experiences (meta-learning). We investigate the combination of seven learning methods and seven image descriptors aiming at creating low-cost classifiers for coffee crops recognition. The objective is to provide an effective mechanism for coffee crop recognition by fusion of region-based classifiers in remote sensing images. The experiments showed that the proposed framework for fusion of classifiers produces better results than the traditional majority voting fusion approach and all base classifiers tested.