Adaptive flame detection using randomness testing and robust features

Autonomous Flame Detection in Videos With a Dirichlet Process Gaussian Mixture Color Model

“…Apart from color and dynamic properties of flames, textures [11], shapes [19] and other features are also employed for flame detection, but mostly together with other properties.…”

Computer vision for wildfire research: An evolving image dataset for processing and analysis

“…Other systems are also exploited. Among these systems are those permitting the extraction of luminance-chrominance components, such as YCbCr [4,41,34],YUV [20] and L * a * b * [3]. There are also work using so-called cylindrical systems such as TSI [15] and TSV [18].…”

“…In this case, it is important to have a database including a large number of heterogeneous fire color images. To complete this part, several fire detection algorithms use motion analysis [41] to consider or delete fire pixel candidates selected using color criteria. Three works present the comparison of the performances of fire pixel detection algorithms on datasets of wildfire images [9,34,38].…”

Implementation and Performance Evaluation of a Video-Equipped Real-Time Fire Detection Method at Different Resolutions using a GPU

“…이와 같은 단 점을 극복하기 위해 CCTV장치 등을 이용한 비디오 기반의 화재감지 방법 연구가 활발히 진행되고 있다 [3][4][5][6][7][8][9][10][11][12][13][14].…”

“…그림 1은 기존 연구에서 각 단계별로 사용했던 방법들을 보여준다. 첫 번째 단계인 색상 분할에서는 RGB(red, green, blue) [3], YCbCr (Y:luminance, Cb: chrominance-blue Cr: chrominance-red) [4][5] 등의 색 공간에서 화재의 특징적인 색상을 이용하여 화재 영역을 분리 하는 연구가 선행되어 왔다. 이 중 YCbCr 색 공간을 이용한 색상 분할은 화재 색상뿐만 아니라 주변과의 휘도를 이용하기 때문에 화재가 가진 색상 특징을 분리하는데 유리하다.…”