Abstract:
위성항법 시스템과 비전 시스템 융합 기술 기반 자동 항법 기술이 개시된다. 본 발명의 일 면에 따른 자기위치 측위장치는 이동체 전방의 영상 획득을 위한 영상 획득부와, GPS(Global Positioning System) 위성으로부터 송신되는 신호를 수신하여 상기 이동체의 위치정보 데이터를 획득하는 위치정보 수신부와, 상기 영상 획득부에서 매 시각(epoch)마다 획득된 영상을 처리하여 객체를 인식하고 인식된 객체와 상기 이동체 사이의 거리를 연산한 비전 데이터를 출력하는 비전 처리부와, 상기 위치정보 수신부와 상기 비전 처리부(30)가 동기화된 데이터를 출력하도록 상기 위치정보 수신부와 상기 비전 처리부 사이의 시간 동기를 수행하는 시간 동기화부를 포함한다. 여기서, 상기 시간 동기화부는 상기 위치정보 수신부의 시간을 기준으로 상기 비전 처리부의 시계를 보정하는 것을 특징으로 한다.
Abstract:
The present invention relates to a device and a method for measuring an initial location coordinate of a moving object. The present invention can measure a location coordinate of the moving object in succession even when it is difficult to measure the location of the moving object, by using a satellite navigation system by measuring the initial location coordinate of the moving object using information between an object recognized through a vision system of the moving object, and the moving object.
Abstract:
Disclosed are a position decision apparatus and a method for same using a vision system. A position decision apparatus using a vision system comprises a determination unit which determines if recognition of an object positioned in n (n is a natural number of 4 or more) number of directions is possible in at least one camera and a processor which, once the recognition is determined to be possible, uses at least one of a first distance against a visible satellite and a second distance against the recognized object so as to determine current location.
Abstract:
The present invention relates to a method for determining a position to reduce the effects of multipath interference from a satellite navigation system, comprising the steps of: determining the distribution of the elevation angles for a plurality of satellites; determining a receiving strength of a signal from each satellite; determining the weight of the signal therefrom based on the distribution of the elevation angles and the receiving strength thereof; and determining the position based on the signal and the weight therefrom. [Reference numerals] (AA) Start; (BB,DD,FF,HH) No; (CC,EE,GG,II) Yes; (JJ) End; (S310) Determine the number of available satellites; (S320) Determine the distribution of the elevation angles; (S330) Determine a receiving strength of a signal; (S340) Number of entire available satellites >= 5; (S350) Receiving strength is excellent?; (S360,S380) Elevation angle is excellent?; (S370) Determine a first weight; (S390) Determine a second weight; (S400) Determine a third weight; (S410) Determine an uniform weight
Abstract:
To solve an inherent problem that a global positioning system has, the present invention provides a device and a method for calculating position for reliable and stable location determination by fusing with a vision system. According to the embodiment, the device and the method for calculating position are able to develop a more improved position determination system by fusing with a vision system to improve continuity and reliability in position determination and solve weaknesses of less accurate position determination due to an obstacle or an external influence and poor continuity due to a shading region. [Reference numerals] (210) Calculation unit; (220) Determination unit; (230) Computation unit; (240) Global positioning system; (250) Vision system; (AA) Processor; (BB) Memory
Abstract:
본 발명은 구조광 패턴 기반의 3차원 영상 정보 추정 장치 및 방법에 관한 것이다. 본 발명에 따른 구조광 패턴 기반의 3차원 영상 정보 추정 장치는 대상체로부터 영상에서의 구조광 패턴을 획득하는 스테레오 카메라 구성이 가능한 복수의 카메라; 및 구조광원부와 상기 복수의 카메라 중 하나의 카메라 간의 화소별 제1 깊이값과, 복수의 카메리로 구성된 스테레오 카메라에서 획득한 양안 영상으로부터 화소별 제2 깊이값을 대상체의 각 화소 영역에 대해 서로 일치하는 화소들의 시차값을 이용하여 추정하고, 추정한 화소별 제1 깊이값에 대한 제1 신뢰도값과 추정한 화소별 제2 깊이값에 대한 제2 신뢰도값을 깊이값 추정 과정에서 사용한 비용 함수를 이용하여 계산하며, 제1 신뢰도값과 제2 신뢰도값을 비교하여 신뢰도값이 높은 깊이값을 선택하여 최종 깊이맵을 획득하는 깊이맵 추정기를 포함한다.
Abstract:
PURPOSE: A detection method of a vehicle on road is provided to increase the accuracy of detection by limiting the power of true-false which is a weakness of algorism for detecting vehicles in existing image field, and to reduce time for treating comparison with case detecting whole images. CONSTITUTION: A detection method of a vehicle on road comprises: a step of inputting image by photographing a front side or a back side of a vehicle through at least one camera installed in a detecting vehicle; a step of detecting vanishing point on road by using image-treating algorism of the input images; a step of setting the region of interest(ROI) based on the vanishing point; and a step of detecting vehicles by image-treating the region of interest.
Abstract:
PURPOSE: A failure object detection apparatus and a method thereof are provided to generate aerial view mapping of an object and to perform aerial view mapping through grouping and clustering. CONSTITUTION: An aerial view mapping unit(210) displays a location of a failure object on a plane. A grouping unit(220) groups the aerial view map according to each row. A clustering unit(230) generates a depth map histogram according to each group. The clustering unit detects a peak and a valley on the histogram. The clustering unit detects the failure object area according to each group.
Abstract:
PURPOSE: A method and an apparatus for performing an image pixel operation are provided to process the summation of mask area pixels through a summation-table operation method based on an integral image. CONSTITUTION: An apparatus receives an input image(S210). The apparatus partitions the input image into a plurality of segmented windows(S230). The apparatus determines a segmented window computation mode based on an inclusion relation of the partial window(S250). The apparatus performs final integral image computation(S270). The apparatus stores an integral image in four buffers(S290).