Bathymetric Lidar utilizes laser and scanning mechanism for measurement. From the recorded return signal of laser pulse, which is affected by both media and boundary between media layers, not only the water depth, but also the substrate information could be collected. This study explores the potential of substrate classification with reflectivity derived from AHAB Hawkeye II bathymetric lidar. Object-based classification scheme is adopted. Besides the reflectivity, eight GLCM based textures, namely, mean, variance, homogeneity, correlation, contrast, dissimilarity, entropy, and angular second moment are also computed. Image segmentation is applied for producing objects and then training areas of four classes are selected. The nearest neighbor classifier is applied for classifying the objects. The result is analyzed with Jeffries-Matusita distance, one index for measuring separability. It is shown that all J-M distances between the classes are larger than 1.710, which is larger than the threshold value of  .

Using MTSAT stalellite images from 2009 to 2012 the spatial and temperal distribution of cloud coverage over Taiwan is investigated in this paper. More cloud coverage over lands than that over the sea by 19.5% in average is analyzed. The conclusion is different from the previous result indicating global cloud coverage over the sea exceeding that over land by 23%. Monthly averaged cloud coverage over the sea and coastal area in July is lowest among one year. Cloud coverage over three areas in the winter is higher than in the summer. Cloud coverage over land from April to June is generally higher than those in other months. Hourly variation of cloud coverage from am. 9 to pm 4 in a day shows that coverage in the noon exceeds those in other hours and that coverage over the sea exceeds that over land in a day.  Gaussian Mixture Model(GMM) and cluster analysis were applied to classifying three groups of cloud coverage over coastal area condersing spatial and temperal distribution. Eastern coast has the highest coverage among three groups; southern as well as southwestern coasts has the lowest one. The results can offer good references to help some researches in other fields, such as remote sensing in coastal area, marine ecology, aquaculture, agriculture and environment of residence.

Asian waters are abundant with nonlinear internal waves. Andaman Sea is one of the nonlinear internal wave active marginal seas. We like to know the spatial distribution of internal waves, the generation and evolution of internal waves in the Andaman Sea. Thus, we collected thousands of SAR images and MODIS images in Andaman Sea region. After compiler all of the images we collected, an internal wave distribution map of Andaman Sea was built up. Numerous of internal wave characters in Andaman Sea are identified such as internal wave sources, wave-wave interaction, wave reflection and internal waves generated in Andaman Sea but propagated outward into India Ocean are found and discussed.

The purpose of this study is to verify the feasibility of using feature tracking method on the images taken by a digital camera near coastal areas. This technology is applied to the surface circulation in the Yin-Yang Bay, a small bay at northeastern Taiwan, with an obvious front of cloudy yellow fresh water and clear brine. The primary images taken in Yin-Yang Bay have to do geometric transformation, image smoothing, and edge detection before working on feature tracking. Two feature tracking methods are applied to these processed images for estimating surface current velocity. One is a subjective method of estimating distance among the sequence images by naked eyes and the other is an objective method of using Maximum Cross Correlation (MCC). The results show that the velocities of surface currents in the Yin-Yang Bay derived from both methods are similar and are comparable to the results of in-situ measurements. This indicates that the technology developed in this study is suitable for coastal circulation research.

This study discusses the detection of thermal plume using X-band radar images. After a series of image processing, we can detect an obvious echo band pattern at the drain outlet of a coastal power plant. We confirm this echo band pattern is induced by the thermal plume after comparing with the simultaneous in-situ sea surface temperature records. We also reveal that the plume pattern is oscillated due to the influence of tidal current. Although the X-band radar image cannot provide the water temperature information within the plume, we can obtain the spatio-temporal features of the thermal plume which is useful for coastal management and environmental protection.

Microwave radar has been applied for nearshore current measurement. But most of them are implemented by a representative current from a whole image which is not reasonable for nearshore area. The purpose of this study is going to derive the spatial current distribution according to image spectral filter method based on dispersion relation. This study measured current at Fuguei Cape and Keelung sea area by microwave radar, and compared with Acoustic Doppler Current Profiler (ADCP) to verify the accuracy. The results confirm the reliance and correctness of current data form radar. The observation from Fuguei Cape found that the averaged current speed is higher than 1 m/s far from 2 km offshore area. The max. current speed observed is 3.92 m/s at Fuguei Cape coast. A strong flow and rip current were found in the east side of the Fuguei Cape and within the bay. They may cause by the interaction between tidal and coastal topography. The observation from Keelung sea area found a counterclockwise vortex one hour before and after low tide with one kilometer radius which occurs in the south-east of Keelung Sill. The strongest speed (1.55 m/s) occurs at 1.5km from Keelung sill with depth of 20 m.