Within the framework of the project, an advanced, autonomous coastal video monitoring system has been developed and installed at Kamari beach (Santorini), aiming to provide continuous records and long-term monitoring of the shoreline position and the swash zone. The system allows the long-term, low-cost and automatic monitoring of crucial beach processes, providing high frequency data which would otherwise require significant allocation of resources both in equipment and personnel. The optical system consists of 2 network cameras, a meteorological station and a field PC for central control and management of the station.
The field station is completely autonomous, while at the same time offers the possibility of remote control of its operational parameters and monitoring of the beach environmental conditions in real time. The optical system is calibrated (lens distortion, cameras height from sea level, ground control points – GCPs) on such a way that the initial raw optical/video data and their products can be converted to georectified timeseries of specialized beach imagery (georectified mosaics).
Flow chart of the coastal video monitoring system installed in Kamari beach in Santorini. The system consists of the autonomous field station, supported by complementary infrastructure at the data server of the Coastal Morphodynamics – Management and Marine Geology Laboratory of the Department of Marine Sciences.
Two network cameras (Vivotek) were installed at 22 m height above the sea level, with a field of view covering the southern section of Kamari beach. The cameras are recording video (3gp) with a 1920 × 1080 pixel resolution and a sampling rate of 5 frames per second. The cameras are fixed, meaning that both their position and field of view remain constant for the entire life span of the station.
Installation and calibration of the optical monitoring system and the meteorological station in Kamari beach.
In order to produce georectified imagery, a series of supplementary data are needed to calibrate the system. Calibration procedures include the geometric correction of the optical data due to lens distortion and the collection of ground control points (GCPs). GCPs match specific image pixels to known geographic coordinates (longitude and latitude), thus allowing the conversion of the raw video frames to georectified optical data for each camera separately.
Kamari beach: Ground Control Points – GCPs for Camera #1, using the proper image geometry after correction of lens distortion; inset images show GCP data collection using an RTK-DGPS.
Left panel: Georectified mosaic of Kamari beach combining images from both cameras; the latter are shown in the insets (25 June 2916, time 13:00). Right panel: Detail of the georectified mosaic; the yellow line denotes the coastline position at 26 June 2016, as measured ( RTK-DGPS).
The basic products of the video monitoring system are the TIMEX and SIGMA composite images from hourly 10-minute recording bursts, which are periodically (and automatically) extracted from the station PC. TIMEX images quantify the time-average of all recorded frames per 10-minute burst, while SIGMAs express the variance (i.e. how much a pixel’s intensity “changes” over the examined time period). For instance, periodical changes in pixel color (between sand and foam) within the swash zone leads to a discrete capture of a high deviation zone/line in SIGMA images. Similarly, other important physical beach parameters such as the wet/dry beach face or the upper limit of the swash zone may be identified/recorded.
Basic products of the optical system for a 10-minute recording period. a-b): ΤΙΜΕΧ, c-d): SIGMA.
Continuous operation of the optical system offers the opportunity of long-term and low cost monitoring of the beach morphodynamic processes; this is also a legal requirement for all future coastal works according to the European Directive 2014/52/EU. In the figure shown below, selected georectified mosaics of Kamari beach (22 December 2016) are shown (the mosaics have been rotated using the cameras position as a center).
Georectified mosaics of Kamari beach at 22 December 2013, Time 09:00.(a) snapshot (b) TIMEX (c)SIGMA and (d) IMMAX (the highest wave run up extent) images.
Shoreline variability is monitored in high spatio-temporal resolution using time series of coastal video images and a fully-automated 2-D shoreline detection algorithm. The detector is based on a localised kernel that progressively ‘walks’ along the feature of interest on the raw or geo-rectified TIMEX imagery, automatically following the high intensity zone along the shoreline.
Automated shoreline detection in georectified SIGMA mosaic (16 December 2016, 16:00). a) SIGMA detail, showing the progressive detection of local maxima (black pixels) within the scanning moving window (grey region). b) Final shoreline detection; the green box at the bottom left part of the SIGMA mosaic denotes the detail in a). c) The same detection shown in a) and b), superimposed on the corresponding TIMEX.
