Makoto Fujiyoshi

Improvement in the performance of camera based vehicle detector for parking lot

The most prevailing approach now for parking lot vehicle detection system is to use sensor-based techniques such as ultrasound and infrared-light sensors. A few engineering firms provide camera-based systems, which are only for underground and indoor parking lots due to the poor accuracy of the detector. The main impediments to the camera-based system in applying to outdoor parking lots are adherent rain drops on the lens in the rain, glaring sun light and dark shadows in the daytime, and low-light intensity and back-lighting in the nighttime. To date, no camera-based detecting systems for outdoor parking lots have been in practical use. This paper reports on the performance of the detector based on the fuzzy c-means (FCM) clustering and the hyperparameter tuning by particle swarm optimization (PSO). The new system was introduced to an underground parking lot in Tokyo in early October 2009 and achieved the detection rate (sensitivity/specificity) of 99.9%. The system was also tested at an outdoor (rooftop) parking lot for a period of two months and achieved 99.6%. The performance clearly surpassed the initial goal of the project. In terms of classification accuracy, the FCM classifier is better than the support vector machine (SVM) and the computation time for training is an order of magnitude smaller than that of SVM.

Vacant parking space detector for outdoor parking lot by using surveillance camera and FCM classifier

The most prevailing approach now for parking lot vacancy detecting system is to use sensor-based techniques. The main impediments to the camera-based system in applying to parking lots on rooftop and outside building are the glaring sun light and dark shadows in the daytime, and low-light intensity and back-lighting in the nighttime. To date, no camera-based detecting systems for outdoor parking lots have been in practical use. A few engineering firms provide the camera-based system, which is only for underground and indoor parking lots. This paper reports on the new camera based system called ParkLotD for detecting vacancy/occupancy in parking lots. ParkLotD uses a classifier based on fuzzy c-means (FCM) clustering and hyper-parameter tuning by particle swarm optimization (PSO). The test result of the detection error rate for the indoor multi-story parking lot has improved by an order of magnitude compared to the current system based on the edge detection approach. ParkLotD demonstrates high detection performance and enables the camera-based system to achieve the practical use in outdoor parking lots.

Using gyroscopic sensors data with artificial neural networks for junction detection

Gyroscopic sensors are frequently used in automotive navigation systems in order to improve location estimation by using the angle information to supplement other sensors such as GPS location data. Gyroscopic sensors can become a major sensor for location estimation for automobiles or autonomous robots for situations where GPS data are inaccurate or cannot be received. In this paper, we assume a situation where GPS data cannot be received, e.g. in a building or tunnel, and gyroscopic sensors and speedometer are the only available sensors for location estimation. We propose applying artificial neural networks to gyroscopic sensor data in order to estimate the current location of the automotive device. We conducted an experiment using an electric model railroad to verify the accuracy of the proposed method.