Knowledge-based tensor subspace analysis system for kinship verification.

Most existing automatic kinship verification methods focus on learning the optimal distance metrics between family members. However, learning facial features and kinship features simultaneously may cause the proposed models to be too weak. In this work, we explore the possibility of bridging this gap by developing knowledge-based tensor models based on pre-trained multi-view models. We propose an effective knowledge-based tensor similarity extraction framework for automatic facial kinship verification using four pre-trained networks (i.e., VGG-Face, VGG-F, VGG-M, and VGG-S). Therefore, knowledge-based deep face and general features (such as identity, age, gender, ethnicity, expression, lighting, pose, contour, edges, corners, shape, etc.) were successfully fused by our tensor design to understand the kinship cue. Multiple effective representations are learned for kinship verification statements (children and parents) using a margin maximization learning scheme based on Tensor Cross-view Quadratic Exponential Discriminant Analysis. Through the exponential learning process, the large gap between distributions of the same family can be reduced to the maximum, while the small gap between distributions of different families is simultaneously increased. The WCCN metric successfully reduces the intra-class variability problem caused by deep features. The explanation of black-box models and the problems of ubiquitous face recognition are considered in our system. The extensive experiments on four challenging datasets show that our system performs very well compared to state-of-the-art approaches.

Feature fusion via Deep Random Forest for facial age estimation.

In the last few years, human age estimation from face images attracted the attention of many researchers in computer vision and machine learning fields. This is due to its numerous applications. In this paper, we propose a new architecture for age estimation based on facial images. It is mainly based on a cascade of classification trees ensembles, which are known recently as a Deep Random Forest. Our architecture is composed of two types of DRF. The first type extends and enhances the feature representation of a given facial descriptor. The second type operates on the fused form of all enhanced representations in order to provide a prediction for the age while taking into account the fuzziness property of the human age. While the proposed methodology is able to work with all kinds of image features, the face descriptors adopted in this work used off-the-shelf deep features allowing to retain both the rich deep features and the powerful enhancement and decision provided by the proposed architecture. Experiments conducted on six public databases prove the superiority of the proposed architecture over other state-of-the-art methods.

Influence of dietary intake of dairy products on dysmenorrhea.

AIM: To determine the frequency of dysmenorrhea and its associated symptoms amongst a number of adolescent female students and to investigate the possible association between daily dairy product intake and dysmenorrhea. METHODS: A self-assessment questionnaire was completed by 127 female university students aged between 19 and 24 years. Participants gave information that included demographics, the nature, type, and severity of pain associated with menstruation if any, management used to relieve dysmenorrhea, associated symptoms, and a general assessment of dietary intake of dairy products. RESULTS: The prevalence of primary dysmenorrhea in the population studied was 87.4% with the majority of the participants' pain symptoms beginning a few days before and continuing through the first two days of menstruation. Forty-six percent of students were found to have severe dysmenorrhea. Abdominal bloating was the most frequently expressed symptom associated with dysmenorrhea amongst the population studied. Dysmenorrhea and associated symptoms were found in significantly fewer female students who consumed three or four servings of dairy products per day as compared to participants who consumed no dairy products. CONCLUSION: Primary dysmenorrhea is common in young women. This study helps us to better understand the relationship between low dietary intake of dairy products and the risk of dysmenorrhea.

Extraction of specific parameters for skin tumour classification.

In this paper, a methodological approach to the classification of tumour skin lesions in dermoscopy images is presented. Melanomas are the most malignant skin tumours. They grow in melanocytes, the cells responsible for pigmentation. This type of cancer is increasing rapidly; its related mortality rate is increasing more modestly, and inversely proportional to the thickness of the tumour. The mortality rate can be decreased by earlier detection of suspicious lesions and better prevention. Using skin tumour features such as colour, symmetry and border regularity, an attempt is made to determine if the skin tumour is a melanoma or a benign tumour. In this work, we are interested in extracting specific attributes which can be used for computer-aided diagnosis of melanoma, especially among general practitioners. In the first step, we eliminate surrounding hair in order to eliminate the residual noise. In the second step, an automatic segmentation is applied to the image of the skin tumour. This method reduces a colour image into an intensity image and approximately segments the image by intensity thresholding. Then, it refines the segmentation using the image edges, which are used to localize the boundary in that area of the skin. This step is essential to characterize the shape of the lesion and also to locate the tumour for analysis. Then, a sequences of transformations is applied to the image to measure a set of attributes (A: asymmetry, B: border, C: colour and D: diameter) which contain sufficient information to differentiate a melanoma from benign lesions. Finally, the various signs of specific lesion (ABCD) are provided to an artificial neural network to differentiate between malignant tumours and benign lesions.

Tissue classification for MRI of thigh using a modified FCM method.

Fuzzy C-means (FCM) has been frequently used to image segmentation in order to separate objects. The most used segmentation attribute is grey level of pixels. Nevertheless, this method can not identify complex image objects because grey level can not take into account all visual information. This paper describes a modified FCM method for tissue classification which integrates separation and fusion operation of partition tree with expert knowledge. Our method has been applied to 26 MRI (Magnetic Resonance Imaging) images of thigh for localizing four main anatomical tissues: muscle, adipose tissue, cortical bone, and spongy bone. A testing dataset of 6500 representative points has been created by an expert. Using our method, we obtain a high classification rate (95.73%) in the test dataset, which largely improved the classification results obtained from existing methods.

Combining WMN and FOCUSS recursive approach to estimating the current density distribution in the brain.

In this paper we present a new approach which combines the two methods of cerebral electric activity's localization: "Weighted Minimum Norm" (WMN) and the iterative method "FOCal Underdetermined System Solver" (FOCUSS). Our idea is to use the current density distribution estimated by the WMN method in order to initialize the weighting matrix necessary for the localization with FOCUSS method. We compare the found results with those of the traditional WMN and FOCUSS methods in term of computing time and resolution matrix. The presented results show that our approach gives a good localization of the active sources in the brain.

[The pterygoclival trihedron: its 3D reconstruction, stability and possible variations during growth]. / Le trièdre ptérygo-clivien: reconstruction en 3D, stabilité ou variations possibles au cours de la croissance.

The growth's study, from skull's radiographics, needs the use of superposition's structures. Three X-ray radiographics pictures from the front, from profile and from under are retalling by a soft. This orthogonalization's step has been realized for every child at two different ages. The reconstruction of the "clivus" straight and "pterygoïde" straights from three views gives the "pterygo-clivus triedre" which stability is studied for time.

[A method of orthogonalization of X-ray negatives by image processing permitting a 3D reconstruction of the elements of the head]. / Méthode d'orthogonalisation de clichés radio X par traitement d'images autorisant la reconstruction 3D d'éléments de la tête.

In order to know better the unrolling of the growth, we dispode several radiographics elements which is sometimes difficult to connect with themselves. Using the computer, may give us the solution to this problem. We describe here a soft which authorizes the 3D reconstruction dating from X-ray radiographic pictures. During step called orthogonalization, the three pictures from the front, from profile and from under, are retalling by the soft. When it's realized, any visible point from two views will automatically be visible in the third view.