Ismael Lengua

Real-time assistance prototype — A new navigation aid for blind people

This paper presents a new prototype for being used as a travel aid for blind people. The system is developed to complement traditional navigation systems such as white cane and guide dogs. The system consists of two stereo cameras and a portable computer for processing the environmental information. The aim of the system is to detect the static and dynamic objects from the surrounding environment and transform them into acoustical signals. Through stereophonic headphones, the user perceives the acoustic image of the environment, the volume of the objects, moving object direction and trajectory, its distance relative to the user and the free paths in a range of 5m to 15m. The acoustic signals represent short train of delta sounds externalized with non-individual Head-Related Transfer Functions generated in an anechoic chamber. Experimental results show that users were able to control and navigate with the system safety both in familiar and unfamiliar environments.

Euro Banknote Recognition System for Blind People

This paper presents the development of a portable system with the aim of allowing blind people to detect and recognize Euro banknotes. The developed device is based on a Raspberry Pi electronic instrument and a Raspberry Pi camera, Pi NoIR (No Infrared filter) dotted with additional infrared light, which is embedded into a pair of sunglasses that permit blind and visually impaired people to independently handle Euro banknotes, especially when receiving their cash back when shopping. The banknote detection is based on the modified Viola and Jones algorithms, while the banknote value recognition relies on the Speed Up Robust Features (SURF) technique. The accuracies of banknote detection and banknote value recognition are 84% and 97.5%, respectively.

Obstacle detectors for visually impaired people

This paper carries out a review on Electronic Travel Aid Systems (ETAS) for visually impaired people and describes a new wearable Cognitive Aid System for Blind People (CASBliP) developed within the frame of European CASBliP project, in which the authors are taking part. Information on the environment enables humans and vertebrates to know about sources that are in many different directions, particularly signals that are outside the detection range of other senses. Sound source localization is inherently important for safety-survival and navigation. In addition to the acoustical cues, the visual cues such as object detection, tracking and distance measurement play an important role in the navigation not only for robots, but also for blind people, since they are often dependent on artificial intelligence. Due to the fact that blind people make maximum use of sound not only to know the obstacle presence, but also how dangerous it is, in order to avoid it effectively, the CASBliP devices use acoustical sounds in order to represent the visual information detected by the sensors and artificial vision systems.

3D CMOS sensor based acoustic object detection and navigation system for blind people

The paper presents a new wearable Cognitive Aid System for Blind People (CASBliP). The prototype device was developed as an obstacle detector, orientation and navigation Electronic Travel Aid (ETA) for blind people. The device provides a binaural acoustic image representation. The environmental information acquisition system is based on an array of 1×64 CMOS Time-of-Flight sensors. Through stereoscopic (binaural) acoustic sounds the device relays the surrounding near and far environment. Experimental results demonstrate that blind users are able to detect obstacles and navigate through unknown and known environments safety and confidently. CASBliP works accurately in range of 15m in distance and 64° in azimuth, providing significant advantages in comparison with currently existing ETA systems.

Material sound source localization through headphones

In the present paper a study of sound localization is carried out, considering two different sounds emitted from different hit materials (wood and bongo) as well as a Delta sound. The motivation of this research is to study how humans localize sounds coming from different materials, with the purpose of a future implementation of the acoustic sounds with better localization features in navigation aid systems or training audio-games suited for blind people. Wood and bongo sounds are recorded after hitting two objects made of these materials. Afterwards, they are analysed and processed. On the other hand, the Delta sound (click) is generated by using the Adobe Audition software, considering a frequency of 44.1 kHz. All sounds are analysed and convolved with previously measured non-individual Head-Related Transfer Functions both for an anechoic environment and for an environment with reverberation.The First Choice method is used in this experiment. Subjects are asked to localize the source position of the sound listened through the headphones, by using a graphic user interface. The analyses of the recorded data reveal that no significant differences are obtained either when considering the nature of the sounds (wood, bongo, Delta) or their environmental context (with or without reverberation). The localization accuracies for the anechoic sounds are: wood 90.19%, bongo 92.96% and Delta sound 89.59%, whereas for the sounds with reverberation the results are: wood 90.59%, bongo 92.63% and Delta sound 90.91%. According to these data, we can conclude that even when considering the reverberation effect, the localization accuracy does not significantly increase.

Face detection and recognition application for Android

This work describes the development of a face detection and recognition application developed into Raspberry Pi and Android. The application connects with the Raspberry Pi by Bluetooth protocols. The object detection is based on boosted cascade while the face recognition is based on Eigenfaces. The developed system may be especially useful for visually impaired users since it can contribute to facilitate their autonomous behavior during their everyday life. The developed device shows great potential for extrapolation to other areas as education of visually impaired users.

Bayesian approach to color-difference models based on threshold and constant-stimuli methods

An alternative approach based on statistical Bayesian inference is presented to deal with the development of color-difference models and the precision of parameter estimation. The approach was applied to simulated data and real data, the latter published by selected authors involved with the development of color-difference formulae using traditional methods. Our results show very good agreement between the Bayesian and classical approaches. Among other benefits, our proposed methodology allows one to determine the marginal posterior distribution of each random individual parameter of the color-difference model. In this manner, it is possible to analyze the effect of individual parameters on the statistical significance calculation of a color-difference equation.

The effect of buffering noise in passenger car

This paper presents the effect of buffering noise in passenger cars in real conditions at variable speed. At 25% of opened back window, the effect of buffering noise is not perceived. When increasing the aperture of the window to 50%, is perceived the creation of the buffering noise effect with resonance frequency = 15.6 Hz at speed = 80 km/h, but is not irritating because the pressure level is low, pressure level = 110.2 dB. At 75% of the window aperture appears two air flows in the inside car: the one coming from the superior part of the superior zone of the inside and another one from the central zone of the inside. The effect of buffering noise is clearly perceived with resonance frequency = 16.4 kHz at speed = 80 km/h and pressure level = 119 dB and 19.5 Hz at speed = 110 km/h and pressure level at 122 dB. At speed higher than 103–140 km/h, the buffering noise disappears, depending on the car model. For Opel Corsa, the effect of buffering noises appears at the same speed range, whereas for Peugeot an...

Time dependence of the time interval in distance/ intensity threshold in human listeners

The paper presents the distance threshold for complex virtual sounds localization in dependence of the time interval. The method of measurement was a forced-choice, three-down one-up staircase. The sound is a complex record of a click sound of 47 ms at 44.1 kHz measured by using a maximum length binary sequence in an anechoic chamber. The azimuth of the experimental is 0°, at the center of the human head where the ITD = 0, varying just the intensity. The investigation reported that the lowest distance thresholds of 0.06 m occurred for high inter click intervals at 200 ms–300ms. At lower ICIs, the distance threshold increases considerably. The experimental was carried out with the objective to find the distance threshold and time interval for implementing the data for a navigation device for blind people.

The effect of buffeting noise in passenger cars

This paper presents the effect of buffeting noise in passenger cars in real conditions at variable speed. When a back window is 25% open the buffeting noise is not perceived; it starts to be perceived when the window is 50% open, with a resonance frequency of 15.6 Hz at a speed of 80km/h, but is not irritating because the pressure level is low (110.2dB). When the window is rolled down at 75%, two airflows appear inside the car: one from the top of the upper interior zone and a second one from the central interior zone. The effect of buffeting noise is then clearly perceived with a resonance frequency of 16.4Hz at a speed of 80km/h and a pressure level of 119dB, increasing to 19.5Hz for a speed of 110km/h and a pressure level of 122dB. At a speed higher than 130-140km/h, the buffeting noise disappears, depending on the car model. For Opel Corsa the effect of buffeting noise appears only in one speed range, while for Peugeot and BMW models three different speed ranges are identified: for generation, persist...