Eigo Segawa

Preceding Vehicle Detection Using Stereo Images and Non-scanning Millimeter-Wave Radar

We developed a system that detects the vehicle driving immediately ahead of ones own car in the same lane and measures the distance to and relative speed of that vehicle to prevent accidents such as rear-end collisions. The system is the first in the industry to use non-scanning millimeter-wave radar combined with a sturdy stereo image sensor, which keeps cost low. It can operate stably in adverse weather conditions such as rain, which could not easily be done with previous sensors. The systems vehicle detection performance was tested, and the system can correctly detect vehicles driving 3 to 50 m ahead in the same lane with higher than 99% accuracy in clear weather. Detection performance in rainy weather, where water drops and splashes notably degraded visibility, was higher than 90%.

Robust Small-Object Detection for Outdoor Wide-Area Surveillance

In this paper, we present a robust small-object detection method, which we call “Frequency Pattern Emphasis Subtraction (FPES)”, for wide-area surveillance such as that of harbors, rivers, and plant premises. For achieving robust detection under changes in environmental conditions, such as illuminance level, weather, and camera vibration, our method distinguishes target objects from background and noise based on the differences in frequency components between them. The evaluation results demonstrate that our method detected more than 95% of target objects in the images of large surveillance areas ranging from 30–75 meters at their center.

Privacy-conscious human detection using low-resolution video

We propose a human detection method for obtaining human flow information from low-resolution video generated by existing surveillance cameras. To use cameras in public spaces, it is necessary to protect the privacy of individuals appearing in the videos. We use low-resolution video in which individuals cannot be identified. In low-resolution video, human detection is more difficult than in high-resolution video because the human region consists of few pixels and has little information. Furthermore, it is challenging to detect an individual when others are captured nearby or occluded the individual. The proposed method offers detection based on the shape of the head, which is kept in a low-resolution image. In addition, to reduce false positives, which occur when detecting heads with simple shapes, head candidates are verified using the shape of the upper-body. The experimental results indicate a detection rate higher than 70% when the head width is from three to eight pixels (corresponding to about 20 to 50 pixels of the human height) in 428 people.