Jacques Facon

Visual speech recognition: a solution from feature extraction to words classification

Audio-visual speech recognition has been an active area of research lately. A bit, and yet unsolved part of this problem is the visual only recognition, or lip reading. Considering an image sequence of a person pronouncing a word, a full image analysis solution would have to segment the mouth area, extract relevant features, and use them to be able to classify the word from those visual features. We approach this problem by proposing a segmentation technique for the lips contours together with a set of features based on the extracted contours which is able to perform lip reading with promising results. We have collected visual speech sequences in our lab and show the results for a set of ten words in Brazilian Portuguese, spoken by different speakers in more than 150 samples. The approach can be extended and applied to other spoken languages as well.

A Fast Hue-Preserving Histogram Equalization Method for Color Image Enhancement using a Bayesian Framework

In this paper, we introduce a new hue-preserving histogram equalization method based on the ROB color space for image enhancement. We use fi-red, G-green, and B-blue 1D histograms to estimate the histogram to be equalized using a Naive Bayes rule. The histogram equalization is performed by shift hue-preserving transformations. Our method has linear time and space complexities, which complies with realtime applications requirements. A subjective assessment comparing our method and other three is performed. Experiments showed that our method is more robust than the others in the sense that neither unrealistic colors nor over-enhancement are produced.

Postal envelope address block location by fractal-based approach

In this paper we propose an approach based on fractal dimension to automatically locate address blocks in postal envelopes. First, the fractal dimension of each pixel of a postal envelope image is computed using the 2D variation procedure. The K-means clustering technique is then used to label pixels as background, noise and semantic objects like stamps, postmarks, and address blocks. A database composed of 200 postal envelope images, with no fixed position for the address block, postmark and stamp is used to evaluate the efficiency of the proposed approach. For each envelope image, the ideal result (ground-truth segmentation) regarding each class has been generated. The comparison between the ground-truth segmentation and the results obtained through the proposed methodology is carried out pixel by pixel. Experiments showed significant and promising results. By using the 2D variation procedure for three ranges of neighbor window sizes (r = {3,5}, r = {3,5,7}, and r = {3,5,7,9}), the proposed approach reached a success rate over 90% on average.

Mathematical morphology and weighted least squares to correct handwriting baseline skew

An approach to correct the baseline handwritten word skew in the image of bank check dates is presented. The main goal of such an approach is to reduce the use of empirical thresholds. The weighted least squares approach is used on the pseudo-convex hull obtained from the mathematical morphology.

Segmentation of postal envelopes for address block location:an approach based on feature selection in wavelet space

This paper presents a segmentation algorithm basedon feature selection in wavelet space. The aim is toautomatically separate in postal envelopes the regionsrelated to background, stamps, rubber stamps, and theaddress blocks. First, a typical image of a postalenvelope is decomposed using Mallat algorithm and Haarbasis. High frequency channel outputs are analyzed tolocate salient points in order to separate the background.A statistical hypothesis test is taken to decide upon moreconsistent regions in order to clean out some noise left.The selected points are projected back to the originalgray level image, where the evidence from the waveletspace is used to start a growing process to include thepixels more likely to belong to the regions of stamps,rubber stamps, and written area. Experiments are runusing original postal envelopes from the Brazilian PostOffice Agency, and here we report results on 440 imageswith many different layouts and backgrounds.

Soccer video segmentation: Referee and player detection

This paper presents a player segmentation methodology for videos of soccer games. This methodology includes some low and high level video processing algorithms, such as dominant color region detection, referee and player identification. Experimental results have shown the efficiency of the methodology in locating and segmenting referee and players.

Lacunarity as a texture measure for address block segmentation

In this paper, an approach based on lacunarity to locate address blocks in postal envelopes is proposed. After computing the lacunarity of a postal envelope image, a non-linear transformation is applied on it. A thresholding technique is then used to generate evidences. Finally, a region growing is applied to reconstruct semantic objects like stamps, postmarks, and address blocks. Very little a priori knowledge of the envelope images is required. By using the lacunarity for several ranges of neighbor window sizes r onto 200 postal envelope images, the proposed approach reached a success rate over than 97% on average.

Skin melanoma segmentation by morphological approach

A new approach to extract the lesion region of melanoma is presented. This approach is based on color morphological operators which are defined from a lexicographic order on the HSI color space. The morphological filtering allows highlighting the region of melanoma that is then segmented by binarization. No a priori knowledge about the process of image acquisition and the type of melanoma is employed and a few heuristics are used. Tests were performed for two sets of benign and malignant melanoma images and compared with the ground-truth lesion segmentation by applying twelve metrics. The results prove the efficiency of this approach with regard to the automatic segmentation of both benign and malignant melanoma.

Bird Species Classification Based on Color Features

This paper presents a novel approach for bird species classification based on color features extracted from unconstrained images. This means that the birds may appear in different scenarios as well may present different poses, sizes and angles of view. Besides, the images present strong variations in illuminations and parts of the birds may be occluded by other elements of the scenario. The proposed approach first applies a color segmentation algorithm in an attempt to eliminate background elements and to delimit candidate regions where the bird may be present within the image. Next, the image is split into component planes and from each plane, normalized color histograms are computed from these candidate regions. After aggregation processing is employed to reduce the number of the intervals of the histograms to a fixed number of bins. The histogram bins are used as feature vectors to by a learning algorithm to try to distinguish between the different numbers of bird species. Experimental results on the CUB-200 dataset show that the segmentation algorithm achieves 75% of correct segmentation rate. Furthermore, the bird species classification rate varies between 90% and 8%, depending on the number of classes taken into account.

Methodology of automatic extraction of table-form cells

The article presents a method for the automatic extraction of the contents of passive and/or active cells in forms. The approach is based on the analysis and recognition of the types of intersection of the lines that make up such cells. Very little a priori knowledge of the form is required. The performance of this approach depends on the correction module mechanisms for detection and correction of errors generated during the intersection identification phase. The potentialities and advantages of this approach are described and illustrated with tests carried out on different form bases.