Jorge Martinez-Carballido

On Signal P-300 Detection for BCI Applications Based on Wavelet Analysis and ICA Preprocessing

This paper describes an experiment on the detection of a P-300 rhythm from electroencephalographic signals for brain computer interfaces applications. The P300 evoked potential is obtained from visual stimuli followed by a motor response from the subject. The EEG signals are obtained with a 14 electrodes Emotiv EPOC headset. Preprocessing of the signals includes denoising and blind source separation using an Independent Component Analysis algorithm. The P300 rhythm is detected through a time-scale analysis based on the discrete wavelet transform (DWT). Comparison using the Short Time Fourier Transform (STFT), and Wigner–Ville Distribution (WVD) indicates that the DWT outperforms the others as an analyzing tool for P300 rhythm detection.

A method to assist in the diagnosis of early diabetic retinopathy: Image processing applied to detection of microaneurysms in fundus images.

Diabetes increases the risk of developing any deterioration in the blood vessels that supply the retina, an ailment known as Diabetic Retinopathy (DR). Since this disease is asymptomatic, it can only be diagnosed by an ophthalmologist. However, the growth of the number of ophthalmologists is lower than the growth of the population with diabetes so that preventive and early diagnosis is difficult due to the lack of opportunity in terms of time and cost. Preliminary, affordable and accessible ophthalmological diagnosis will give the opportunity to perform routine preventive examinations, indicating the need to consult an ophthalmologist during a stage of non proliferation. During this stage, there is a lesion on the retina known as microaneurysm (MA), which is one of the first clinically observable lesions that indicate the disease. In recent years, different image processing algorithms, which allow the detection of the DR, have been developed; however, the issue is still open since acceptable levels of sensitivity and specificity have not yet been reached, preventing its use as a pre-diagnostic tool. Consequently, this work proposes a new approach for MA detection based on (1) reduction of non-uniform illumination; (2) normalization of image grayscale content to improve dependence of images from different contexts; (3) application of the bottom-hat transform to leave reddish regions intact while suppressing bright objects; (4) binarization of the image of interest with the result that objects corresponding to MAs, blood vessels, and other reddish objects (Regions of Interest-ROIs) are completely separated from the background; (5) application of the hit-or-miss Transformation on the binary image to remove blood vessels from the ROIs; (6) two features are extracted from a candidate to distinguish real MAs from FPs, where one feature discriminates round shaped candidates (MAs) from elongated shaped ones (vessels) through application of Principal Component Analysis (PCA); (7) the second feature is a count of the number of times that the radon transform of the candidate ROI, evaluated at the set of discrete angle values {0°, 1°, 2°, …, 180°}, is characterized by a valley between two peaks. The proposed approach is tested on the public databases DiaretDB1 and Retinopathy Online Challenge (ROC) competition. The proposed MA detection method achieves sensitivity, specificity and precision of 92.32%, 93.87% and 95.93% for the diaretDB1 database and 88.06%, 97.47% and 92.19% for the ROC database. Theory, results, challenges and performance related to the proposed MA detecting method are presented.

An FPGA-based architecture for linear and morphological image filtering

Field Programmable Gate Array (FPGA) technology has become a viable target for the implementation of real time algorithms suited to video image processing applications. The unique architecture of the FPGA has allowed the technology to be used in many applications encompassing all aspects of video image processing. Among those algorithms, linear filtering based on a 2D convolution, and non-linear 2D morphological filters, represent a basic set of image operations for a number of applications. In this work, an implementation of linear and morphological image filtering using a FPGA NexysII, Xilinx, Spartan 3E, with educational purposes, is presented. The system is connected to a USB port of a personal computer, which in that way form a powerful and low-cost design station. The FPGA-based system is accessed through a Matlab graphical user interface, which handles the communication setup. A comparison between results obtained from MATLAB simulations and the described FPGA-based implementation is presented.

FPGA-based educational platform for real-time image processing experiments

In this paper, an implementation of an educational platform for real‐time linear and morphological image filtering using a FPGA NexysII, Xilinx®, Spartan 3E, is described. The system is connected to a USB port of a personal computer, which in that way form a powerful and low‐cost design station for educational purposes. The FPGA‐based system is accessed through a MATLAB graphical user interface, which handles the communication setup and data transfer. The system allows the students to perform comparisons between results obtained from MATLAB simulations and FPGA‐based real‐time processing. Concluding remarks derived from course evaluations and lab reports are presented.

Using adaptive threshold to detect pedestrians crossing on a street for Advanced Driver Assistance Systems

Advanced Driver Assistance Systems (ADAS) are becoming of major relevance for driver assistance. Pedestrian Protection Systems are part of ADAS. It is well known that driver distraction on his/her driving is a major cause of traffic accidents, and giving the environment variability while driving it is a challenging area of research. This work shows that by reducing the search space to critical areas, not requiring background elimination, not ROI detection and selection, and as a result simplifying the algorithm for pedestrian detection giving a good approach. The algorithm parameterizes image resolution, thus allowing for different camera resolutions that surely will continue to improve. This work uses at its core the fact that a pedestrian has at least a shoe in contact with the pavement, together with the alarm zones to assist the driver is enough information for an ADAS to decide on actions for driver assistance. On a test set of 46 test images that were taken on a real urban environment this work has results that show an 89% correct detection on a real outdoor environment.

License plate digit recognition using 7×5 binary templates at an outdoor parking lot entrance

Digit recognition for license plate recognition is a core process, simplicity on the algorithm to recognize digits on the LPR context is of importance, given that there is an ever increasing vehicle flow and need for recognizing license plates on different environments and applications such as: unattended parking lots, automatic toll collection, traffic supervision, access control of restricted office and housing areas.

Real Time Human Action Recognition Using Full and Ultra High Definition Video

Given that camera products are moving rapidly towards Ultra High Definition in video surveillance area, this work presents a real-time human action recognition method based on simple and clear concepts which achieves a great Accuracy-Speed performance on videos up 8K Ultra High Definition. The method was evaluated on i3DPost dataset at different video resolutions. The comparative analysis shows that our method outperforms all state of the art methods, in accuracy (99%) and to the best of our knowledge, it is the first one that shows real-time performance on videos up 8K UHD.

Strategies for energy management in industrial microgrids with a local cogeneration approach

In the integration of alternative energy in the electric power system, the microgrids concept provides an adequate platform for massive incorporation of this kinds of energy. This paper presents the multi-objective problem in the microgrids energy management and sets up strategies focused on industrial microgrids based on the electricity market framework and a local generation approach aims to beneficiate the users, the utilities and the environment.

Chirp signal detection using the Duffing oscillator

The chaotic oscillator has become an important tool in the analysis of harmonic signals with low signal to noise ratio. On the other hand, traditionally, frequency modulated (FM) signals have always been studied through conventional techniques of time-frequency analysis. Recently, the Duffing oscillator has shown to be a powerful tool to detect frequencies for periodic signals because of its chaotic properties. It is for this reason that this paper presents a chirp signal analysis in the presence of noise using a Duffing oscillator. The oscillators chaotic characteristic provides information on how the frequency changes in a signal therefore, for the chirp case it is also convenient to apply time-frequency techniques to obtain the frequency shift for such signals. Here, it is found that is possible to use the Duffing oscillator for the analysis of chirp signals with -15 dB of SNR. The time-frequency techniques used in this paper are the Short Time Fourier Transform, the Wigner-Ville Distribution, its corresponding Choi-Williams and end with a proposed array of Duffing oscillators.

Metamyelocyte Nucleus Classification Using a Set of Morphologic Templates

Laboratory tests and clinical analysis is a must to confirm or reject some medical diagnosis. Common analytical fields of knowledge are: Hematology, Clinical biochemistry, Urology, Microbiology among others. This work is for the clinical field of Hematology where detection of immature leukocytes on peripheral blood sample is part of a clinical test of relevance for detection of certain types of cancer. Hematologists distinguish metamyelocytes by their characteristic kidney shape of its nucleus. A novel approach for metamyelocyte classification uses a set of morphological templates. By transforming a pixel based contour to a contour region, morphological templates, the classification becomes a matter of sequence searching on a symbolic string. Using a ten image set for testing, the results are 100% correct. This is very promising, but further extensive set of images could lead to adding and/or modifying morphological templates.