Laura Caponetti

Computer-aided simulation for bone surgery

A system for evaluating bone deformities using a 3-D model directly recovered from 2-D images and for simulating surgery is described. It derives a 3-D object representation from only two X-ray images. It also offers user-friendly simulation of bone surgery with low-cost hardware and software. The system exhibits satisfactory behavior for reconstructing the bone shape, providing suitable data for the simulation and evaluation of bone surgery. Although the spline interpolation of the bone surface does not produce a realistic 3-D visualization of the tibia, which is used as an example, the reconstruction is useful in solving problems inherent in the pathology considered.<<ETX>>

Document page segmentation using neuro-fuzzy approach

In this work, we propose a new document page segmentation method, capable of differentiating between text, graphics and background, using a neuro-fuzzy methodology. Our approach is based firstly on the analysis of a set of features extracted from the image, available at different resolution levels. An initial segmentation is obtained by classifying the pixels into coherent regions, which are successively refined by the analysis of their shape. The core of our approach relies on a neuro-fuzzy methodology, for performing the classification processes. The proposed strategy is capable of describing the physical structure of a page in an accurate way and proved to be robust against noise and page skew. Additionally, the knowledge-based neuro-fuzzy methodology allows us to understand the classification mechanisms better, contrary to what happens when other kinds of knowledge-free methods are applied.

A genetic approach to edge detection

We propose to use optimization techniques known as genetic algorithms in the searching of the optimal edge configuration. An objective function is supplied which assigns a fitness value to each edge configuration. Fitness represents the ability of an individual to survive and reproduce; in our case it is evaluated on the basis of the local nature of the edges.<<ETX>>

Fuzzy mathematical morphology for biological image segmentation

Due to the imaging devices, real-world images such as biological images may have poor contrast and be corrupted by noise, so that regions in the images present soft edges and their segmentation turns out to be quite difficult. Fuzzy mathematical morphology can be successfully applied to segment biological images having such characteristics of vagueness and imprecision. In this work we introduce an approach based on fuzzy mathematical morphology to segment images of human oocytes in order to extract the oocyte region from the entire image. The approach applies fuzzy morphological operators to detect soft edges in the oocyte images, followed by morphological reconstruction operators to isolate the oocyte region. The main concepts from fuzzy mathematical morphology are briefly introduced and the results of applying fuzzy morphological operators are reported in low-contrast images of human oocytes.

Biologically inspired emotion recognition from speech

Emotion recognition has become a fundamental task in human-computer interaction systems. In this article, we propose an emotion recognition approach based on biologically inspired methods. Specifically, emotion classification is performed using a long short-term memory (LSTM) recurrent neural network which is able to recognize long-range dependencies between successive temporal patterns. We propose to represent data using features derived from two different models: mel-frequency cepstral coefficients (MFCC) and the Lyon cochlear model. In the experimental phase, results obtained from the LSTM network and the two different feature sets are compared, showing that features derived from the Lyon cochlear model give better recognition results in comparison with those obtained with the traditional MFCC representation.

A Texture-Based Image Processing Approach for the Description of Human Oocyte Cytoplasm

The purpose of this paper is to develop a diagnostic tool that can analyze light microscope images of human oocytes and derive a description of the oocyte cytoplasm that is useful for quality assessment in assisted insemination. The proposed approach includes three main phases: 1) segmentation; 2) feature extraction; and 3) clustering. In the segmentation phase, a region of interest inside the cytoplasm is extracted through morphological operators and the Hough transform. In the second phase, regions that result from segmentation are processed through a multiresolution texture analysis to extract a set of features that describe different levels of cytoplasm granularity. To this aim, we evaluate some statistics in the Haar wavelet transform domain. Finally, the extracted features are used to cluster oocytes according to different levels of granularity. This approach is made by fuzzy clustering. Experimental results on a collection of microscope images of oocytes are reported to show the effectiveness of the proposed approach. In addition, comparison with alternative methods for feature extraction and clustering is performed.

Fuzzy classification of image pixels

We present a neuro-fuzzy approach for classification of image pixels into three classes: contour, regular or texture points. Exploiting the processing capabilities of a neural network, fuzzy classification rules are derived by learning from data and applied to classify pixels in grey-level images. To derive a proper set of training data, the spatial properties of the image features and a multiscaled representation of images are considered. The effectiveness of the proposed approach is illustrated on some sample images.

Multiresolution texture analysis for human oocyte cytoplasm description

In this work we present an approach based on image texture analysis to obtain a description of oocyte cytoplasm which could aid the medical expert in the selection of oocytes to be used for assisted insemination. More specifically, we describe some characteristics such as different levels of uniformity and/or granularity in the oocyte cytoplasm, using multiresolution texture analysis applied to light microscope images. To this aim, we evaluate some statistical measures in the wavelet transform domain of image regions and classify them according to different levels of granularity. Preliminary experimental results on a collection of light microscope images of oocytes are reported to show the effectiveness of the proposed approach.

Speech emotion recognition using spiking neural networks

Human social communication depends largely on exchanges of non-verbal signals, including non-lexical expression of emotions in speech. In this work, we propose a biologically plausible methodology for the problem of emotion recognition, based on the extraction of vowel information from an input speech signal and on the classification of extracted information by a spiking neural network. Initially, a speech signal is segmented into vowel parts which are represented with a set of salient features, related to the Mel-frequency cesptrum. Different emotion classes are then recognized by a spiking neural network and classified into five different emotion classes.

Content-based recognition of musical instruments

A method for content-based audio classification is presented. In particular we focus on identification of musical instruments sounds based on timbre classification, using a biologically plausible features extraction technique called cochleagram, and a new model of recurrent neural network called LSTM. Preliminary experiments are performed to compare various feature sets and neural network sizes. In particular two experiments are performed, using two different feature sets. The best classification rate obtained is 80%, averaged on 20 trials.