Real-time analysis of fuel spray images

This paper provides a fast and efficient technique for real-time preprocessing and analysis of fuel spray images. The preprocessing stage consists of global thresholding of the log-edge of the image. The recognition of in-focus droplets is achieved through a 3-level tree classifier. The algorithm was tested using 400 images including approximately 8000 candidate objects. The percentage of correct classification was > 93%. The algorithm was also tested using 200 images of very low quality, and a recognition rate of 87% was achieved.