FOREGROUND REGION EXTRACTION DEVICE AND FOREGROUND REGION EXTRACTION METHOD

摘要

According to one embodiment, a foreground region extraction device (100) includes a processor (30) and a detector (50). The processor (30) acquires information of first distances at a first time between a first position (P±) and a first region (R1). The processor (30) acquires information of second distances at a second time between a second positon (P²) and a second region (R2). The processor (30) derives a first image of the first region (R1) projected onto a first plane based on the first distances, and calculates a first feature based on the first image. The processor (30) derives a second image of the second region (R2) projected on to a second plane based on the second distances and a relative arrangement of the first position (R1) and the second position (R2) and calculates a second feature based on the second image. The detector (50) detects a moving body based on the first feature and the second feature.