Fuzzy-Based Automatic Approach for Underwater Docks´ Anomalies Detection

Survey of underwater docks is a necessary task to assure their maintenance that nowadays is carried out by specially trained divers. Inspection of underwater docks´ walls should be carried out at short intervals, however, due to the implicit difficulties of the task and the high cost of the divers´ training process, the surveying work can´t be carried out as often as it would be desirable. As big anomalies can cause structural dock damages, don´t visible to the naked eye but affecting the whole wall, the automation of this task, considering the brightness, lightness, and shadowing problems would allow a more frequently surveillance falling into a more effective and cheaper underwater docks´ maintenance. In this paper we present the results of our ongoing work on automatic visual monitoring system for detecting the anomalies of underwater docks based on color information. As color-order systems based on perceptual variables are somehow correlated with human being´s color perception, the proposed algorithm is based on the HSI Smiths´ color system and makes use of fuzzy techniques to face up the problems due to low saturation and illumination