Mitsubishi Electric Corporation

Mengxiong Wang

Yoshimi Moriya

Akira Minezawa

Hidetoshi Mishima

Hideaki Maehara

Momoyo Hino

Naohiro Shibuya

Kazuo Sugimoto

Kazuyuki Miyazawa

Kenji Taira

Takehiko Hanada

Tetsuro Wada

Tomoya Sawada

Eiji Ueda

An object of the present invention is to provide a water level measurement device and a shoreline extraction method each of which is capable of stably measuring the water level. A pixel selection unit () selects a pixel of interest () from a designated area () designated from a captured image (), and an identification image extraction unit () extracts identification images (), () each coming in contact with the pixel of interest (). The identification unit () calculates an identification strength indicating a degree to which an area corresponding to each of the identification images (), () is a water area, on a basis of the result of machine learning related to identification between the water area and a non-water area. The shoreline extraction unit () extracts a shoreline in the captured image from the identification strength of the area corresponding to each of the identification images (), (), on a basis of the result of machine learning related to the identification between the water area and the non-water area by the learning unit ().

Water level measurement device and shoreline extraction method

A congestion degree estimating unit () which estimates, for each image frame, a congestion degree of an area in which each object detected by an object detecting unit () is present is provided. An object selecting unit () refers to an estimation result of the congestion degree by the congestion degree estimating unit () for each image frame, and selects an object present in an area the congestion degree of which is equal to or lower than a first threshold out of one or more objects detected by the object detecting unit ().

Object tracking device and object tracking method

An image feature map generating unit () generates, on the basis of feature amounts extracted from a plurality of images successively captured by a camera (), an image feature map which is an estimated distribution of the object likelihood on each of the images. An object detecting unit () detects an object on the basis of the image feature map generated by the image feature map generating unit ().

Object detection device and object detection method

A number-of-operations information storage unit () stores number-of-operations information in which a plurality of conditions of the vehicle is defined and number of acceleration operations and number of deceleration operations performed by a vehicle are described for each condition defined. A correction speed specifying unit () detects a current condition of the vehicle, when the vehicle travels at a constant speed, extracts number of acceleration operations and number of deceleration operations described for a condition corresponding to the detected current condition of the vehicle from the number-of-operations information, compares the extracted number of acceleration operations with the extracted number of deceleration operations, specifies a correction speed higher than a set speed for a constant-speed travel if the number of acceleration operations is larger than the number of deceleration operations, and specifies a correction speed lower than the set speed if the number of deceleration operations is larger than the number of acceleration operations. A constant-speed travel control unit () makes the vehicle travel at a constant speed based on the correction speed specified by the correction speed specifying unit ().

Vehicle constant-speed travel control apparatus, vehicle constant-speed travel control method, and computer readable medium

Growing community of inventors

Tokyo, Japan

Mengxiong Wang