Clarion Co

Junya Nakamura

Naoki Shimizu

Takehito Ogata

Kimiyoshi Machii

Ayano Enomoto

Shinji Matsumoto

Makoto Uemura

Kensuke Yamamoto

Yoshikazu Sakai

A section line recognition device is realized which can determine a line type of a section line in a road surface with high accuracy. In step, a feature amount is extracted from an image captured by each camera. This step corresponds to a process of a feature point extraction unit. In the next step, the extracted feature amount is transformed into bird's-eye view coordinates which is common coordinates. This step corresponds to a process of a coordinates transformation unit. Next, in step, a camera to be selected is determined. This step is a process of a camera determination unit. Next, in step, the state transition at the appearance position of the feature point on the bird's-eye view coordinates is determined using feature point coordinates of the camera selected in step. This step is a process of a state transition determination unit. Finally, in step, the line type is determined. This step is a process of a line type determination unit.

Section line recognition device

The present invention addresses the problem of enabling an accurate determination of line type using a plurality of cameras. A demarcation line recognition device of the present invention includes a plurality of cameras (); a line type determination unit () that senses demarcation line candidates from respective images obtained by the plurality of cameras and determines line types of the sensed demarcation line candidates; a demarcation line position calculation unit that calculates the positions of the sensed demarcation line candidates; a reliability level calculation unit that computes a reliability level of each demarcation line candidate by using positional relationships between the sensed demarcation line candidates; a probability computation unit () to which line types of demarcation line candidates for which the reliability level is equal to or greater than a threshold value are input and that determines the probability of each line type; and a line type decision unit () that finally decides the line types of the demarcation lines on the basis of the results of the probability computation unit.

Demarcation line recognition device

Conventional navigation devices are equipped with technology that supports the driving of a driver by detecting the surroundings of a vehicle using a plurality of sensors such as cameras. However, because complex arithmetic processing is carried out using the detection results from the plurality of sensors, the processing load increases and other processes requiring a rapid response can be delayed. As a result, consideration has been given to decreasing the amount of arithmetic processing and reducing the processing load by uniformly decreasing the processing area for information detected by the sensors. However, if the processing area is uniformly decreased, information necessary for driving support may not be obtained. Provided is a sensor control technology that better obtains information necessary for driving support: a sensor controller, characterized by carrying out predetermined processing by changing the processing area of the information acquired by the sensors according to the vehicle driving conditions.

Sensor controller, navigation device, and sensor control method

Provided is a technology capable of improving the accuracy of determining a position of an own vehicle in the vicinity of a branch point, in a current position determining device. The current position determining device identifies a detection range according to states of a road on which a vehicle is traveling to detect a lane change, to thereby determine a current position of the vehicle based on information on the detected lane change.

Current position determining device and current position determining method for correcting estimated position based on detected lane change at road branch

Growing community of inventors

Saitama, Japan

Junya Nakamura