Day and Night Band (DNB) of Visible Infrared Imaging Radiometer Suite (VIIRS) on the Suomi National Polar-Orbiting Partnership (S-NPP) has exhibited a capability to detect the fishery lights in the night time. The distribution of fishery lights, which are corresponding to the distribution of fishery resources, has a significant information for fishery industries and resource managements. Unfortunately, it is difficult to distinguish the fishery lights on DNB images because the lunar lights reflected by clouds are observed simultaneously. In this study, the artificial neural network (ANN) was developed to detect the fishery lights apart from the lunar lights reflected by clouds. The ANN was trained to simulate the DNB lights with the brightness temperature at 3.7μm (BT_3.7) and the fraction of Moon illumination in date and location. The fishery lights were given as the absolute error between the predicted and the observed DNB. The errors were compared between the pixel based ANN and the convolutional NN (CNN), and the pixel based ANN was superior to the CNN due to convolution of the cloud pixels.

In this paper, we examine the terrain effect into terrestrial albedo estimation. Terrestrial albedo is one of the most important parameters in terms of understanding global heat balance. The existing approach for estimating terrestrial albedo starts with the estimation of parameters of bidirectional reflectance distribution function (BRDF) model from measurements observed at different geometry. Then, narrowband albedos are estimated from the BRDF model parameters, and finally the broadband albedo is estimated via narrowband-to-broadband conversion. The previous researches do not consider the terrain effect for generating the terrestrial albedo. The experiments using in-situ measurements show that the BRDF model transforming the geo-coordinate of the reflectance of the shadowed terrain generates the best accuracy. The improvement of the accuracy by the terrain effect correction is limited, and thus it is concluded that further investigation using more in-situ data and simulated data is necessary for operational products.

A visual system consisting of a RGB camera with a laser scanner needs accurate measurement of extrinsic parameters. Whereas, calibration of a camera-laser system is still a fundamental and challenging task. In this letter, we propose a novel calibration method utilizing a typical LCD (Liquid Crystal Display) panel. A pre-experiment revealed that a LCD panel varies laser reflection intensity in response to its brightness. Our method can calibrate extrinsic parameters between a camera and a laser scanner automatically without special devices. An evaluation using laser scanning data and aerial images acquired by flying an unmanned aerial vehicle showed that our calibration method is useful for construction surveying.

As our previous work, we have reported a method to measure the river water level by capturing water gauge images of CCTV cameras for river monitoring into a computer and collating them with one taken at the time of low water level. For this time, we examined a method of measuring the river water level by identifying the water border position based on the deep learning technology, targeting the non-installation sites of the water gauge. We also have estimated the effectiveness of the new method using actual river image data set.