Naval ship recognition

Abstract. Object recognition is a very interesting task with multiple applications and for that reason it has been dealt with very intensively in the last years. In particular, the application to naval ship pictures may facilitate the work of the coastguards or the navy. However, this type of images entails some difficulties due to their specific environment. Water reflects the light and as a consequence, some areas may presumably show different brightness and color. Waves from wind or moving ships pose a problem due to the additional edges that they produce. The camouflage of ships in the military context is also an issue to take into account. Therefore, it is difficult to propose a simple method that is valid for every image.

A discussion about which techniques may solve these problems is presented and finally a combined solution based on contour recognition is suggested. Test images are preprocessed by histogram stretching. Then, the Canny method is applied to the image and to the reference contour in order to obtain not only their edges, but also their respective orientations. The problem of recognizing the reference contour within the detected edges is addressed by making use of the Generalized Hough Transform (GHT).