Abstract:
본 발명은 휘도 및 조명마스크의 밝기에 따른 가중치를 이용한 영상의 콘트라스트 보정 장치 및 방법에 관한 것으로, 보다 구체적으로는 입력 영상에 대해 색공간을 변환하여 휘도신호와 색차신호로 분리한 후, 휘도신호에 대해 영상의 밝은 정도에 따라서 서로 다른 가중치를 부여하는 새로운 렉티넥스 처리에 의해 영상의 밝은 영역 및 어두운 영역의 콘트라스트가 균형있게 향상되게 하며, 원래의 영상과 레티넥스 처리된 영상을 가중합하여 원래의 영상의 특성을 반영할 수 있는 휘도 및 조명마스크의 밝기에 따른 가중치를 이용한 영상의 콘트라스트 보정 장치 및 방법에 관한 것이다.
Abstract:
PURPOSE: A contrast correction device of an image and a method using a weighted value corresponding to brightness of luminance and a lamp mask are provided to increase illumination of a bright area and a dark area. CONSTITUTION: An input color space conversion unit(110) converts a color space of an input image into a color space formed by a luminance signal and a chrominance signal. A lamp mask production unit(120) generates a lamp mask from input luminance signals. An illumination correction processing unit(130) generates an output luminance signal which illumination is corrected. A chrominance signal correction unit(140) generates an output chrominance signal with corrected chroma. [Reference numerals] (110) Input color space conversions unit; (120) Lamp mask production unit; (121) Low-pass filter processing module; (122) Weighted value processing module; (130) Illumination correction processing unit; (131) First illumination correction processing unit; (132) Second illumination correction processing unit; (140) Chrominance signal correction unit; (150) Output color space conversion unit; (AA) First brightness signal; (BB) Second brightness signal; (CC) Output brightness signal; (DD) Input image; (EE) Input brightness signal; (FF) Input chrominance signal; (GG) Output image
Abstract:
A high speed region merger method of a watershed extracting exact object information and an apparatus thereof are provided to merge region at high speed without repetitive merge. An average brightness value of pixel values is evaluated in local regions(100) divided into local watershed. The difference of the average intensity with the domain frame is calculated. One and six are determined as a decision region in a background region. Four and eight are determined as the decision region in an object region. The difference of the average intensity of the domain frame is inputted to a hierarchical queue(102) in the numerical order of localized regions. An altitude(103) of the hierarchical queue is determined based on the difference of the average intensity by the value inputted to the hierarchical queue.
Abstract:
본 발명은 전달률 하한치를 이용한 안개영상 개선 장치에 관한 것으로서, 안개, 연기, 구름 등 공기 중의 밝기 및 색상 성분과, 객체의 빛과 색 성분이 혼합되어 가시성이 낮아진 영상에 대해 공기 중의 밝기와 색상성분을 제거하여 선명한 영상을 제공하도록 함에 그 목적이 있다. 이러한 목적을 달성하기 위한 본 발명은, 안개낀 입력영상을 입력받아 전달률 하한치 및 전달률 추정치를 계산하고, 계산된 전달률 추정치와 하한치를 가중합하여 전달률을 계산하는 전달률 계산부; 상기 전달률 계산부를 통해 계산된 전달률과 안개낀 입력영상을 입력받아, 입력된 상기 전달률, 안개낀 입력영상 및 대기의 밝기값을 이용하여 안개가 제거된 영상을 출력하는 복원영상 계산부; 및 상기 복원영상 계산부를 통해 출력된 영상에 대하여 명도 확장 처리를 수행하여, 최종적으로 안개가 제거된 영상을 출력하는 후 처리부; 를 포함한다.
Abstract:
PURPOSE: A major body part estimating method which uses a directional technician and a machine learning for a posture recognition is provided to accurately estimate a major body part even in case the major body part is overlapped with other body parts by obtaining an outer edge and an inner edge and estimating the major body part by using a human image which has depth information. CONSTITUTION: A program of a computer device receives a human image (S1000). The program detects the edge of the human image (S2000). The program detects end points of a major body part from a generated edge image (S3000). The program detects predicted points of the major body part (S4000). The program obtains the central point of the detected predicted points as a candidate point of the major body part (S5000). The program obtains the edge distance information of the candidate point (S6000). The program estimates the body part of the candidate point by a comparison of the candidate point and learning data (S7000). [Reference numerals] (AA) Start; (BB) End; (S100) Generate learning data by using feature of head, hand and foot; (S1000) Input human image; (S2000) Detect image edge; (S3000) Detect end points of a major body part; (S4000) Detect predicted points of the major body part; (S5000) Obtain candidate point of the major body part; (S6000) Obtain the edge distance information of the candidate point; (S7000) Estimate body part of the candidate point by a comparison of the candidate point and learning data
Abstract:
PURPOSE: A system for providing location tracking service information using a mobile fingerprint recognition terminal and a method thereof are provided to input a fingerprint of a user at a fixed time interval, thereby authenticating the user. CONSTITUTION: A location tracking service server(200) requests fingerprint input to a portable terminal(100) at a fixed time interval. The portable terminal authenticates a user using an inputted fingerprint image. The portable terminal transmits a user authentication result to the location tracking service server. The location tracking service server transmits the user authentication result with location information to a request terminal(300). [Reference numerals] (400) Wireless network
Abstract:
PURPOSE: A method for detecting a face area of a speaker in a communication terminal and the communication terminal using the same, and a method for measuring the distance between speakers and a communication terminal using the same are provided to supply an optimized distance for guaranteeing high communication quality to a speaker. CONSTITUTION: A transceiving unit of a communication terminal receives a k-numbered image frame(S210). A converting unit converts the k-numbered image frame into color signal components corresponding to a plurality of color coordinate systems(S230). A probability mask assigning unit distinguishes the color signal components(S240). A detection unit calculates horizontal and vertical histogram values(S270). The detection unit detects a face area using the histogram value(S280).