Supporting Pedestrians with Visual Impairment During Road Crossing: A Mobile Application for Traffic Lights Detection
Many traffic lights are still not equipped with acoustic signals. It is possible to recognize the traffic light color from a mobile device, but this requires a technique that is stable under different illumination conditions. This contribution presents TL
- PDF / 301,132 Bytes
- 4 Pages / 439.37 x 666.142 pts Page_size
- 81 Downloads / 223 Views
EveryWare Lab, Department of Computer Science, Universit` a degli Studi di Milano, Milan, Italy [email protected] 2 EveryWare Technologies, Milano, Italy
Abstract. Many traffic lights are still not equipped with acoustic signals. It is possible to recognize the traffic light color from a mobile device, but this requires a technique that is stable under different illumination conditions. This contribution presents TL-recognizer, an application that recognizes traffic lights from a mobile device camera. The proposed solution includes a robust setup for image capture as well as an image processing technique. Experimental results give evidence that the proposed solution is practical. Keywords: Blind people · Visual impairments · Mobile device · Smartphones · Traffic lights · Computer vision
1
Introduction
Independent mobility involves a number of challenges for people with visual impairment or blindness, including being aware of the presence and the current color of traffic lights. This is particularly challenging when traffic lights are not equipped with acoustic signals. A number of solutions have been proposed in the scientific literature to recognize traffic lights (among others, [1–3]). Existing solutions share a common problem: they use images acquired through the device camera with automatic exposure. With this approach, in conditions of low ambient light (e.g., at night) traffic lights result overexposed while in conditions of high ambient light (e.g., direct sunlight) they are underexposed (see Fig. 1(a)). This contribution, extracted from our previous work [4], presents TLrecognizer , a software module that addresses the above problem with an effective solution: besides image processing and recognition, it proposes a robust setup for image capture that allows to acquire clearly visible traffic light images regardless of ambient light variability due to time and weather. The proposed recognition technique adopting this approach is reliable (full precision and high recall), robust (works in different illumination conditions) and efficient (it can c Springer International Publishing Switzerland 2016 K. Miesenberger et al. (Eds.): ICCHP 2016, Part II, LNCS 9759, pp. 198–201, 2016. DOI: 10.1007/978-3-319-41267-2 27
A Mobile Application for Traffic Lights Detection
(a)
199
(b)
Fig. 1. Image acquisition with automatic exposure and with our technique in different illumination conditions. (a) Acquisition with automatic exposure: high ambient light (left) and low ambient light (right). (b) Acquisition with our technique: high ambient light (left) and low ambient light (right).
run several times a second on commercial smartphones). The experimental evaluation conducted with visual impaired subjects shows that the technique is indeed practical in supporting road crossing.
2
The Technique to Recognize Traffic Lights
The recognition process is organized in two main phases: ‘input-acquisition’ and ‘image-processing’. During input-acquisition a frame is captured by the device camera using specifically designed exposure parameters. Th
Data Loading...
