Noise2Grad: Extract Image Noise to Denoise

In many image denoising tasks, the difficulty of collecting noisy/clean image pairs limits the application of supervised CNNs. We consider such a case in which paired data and noise statistics are not accessible, but unpaired noisy and clean images are easy to collect. To form the necessary supervision, our strategy is to extract the noise from the noisy image to synthesize new data. To ease the interference of the image background, we use a noise removal module to aid noise extraction. The noise removal module first roughly removes noise from the noisy image, which is equivalent to excluding much background information. A noise approximation module can therefore easily extract a new noise map from the removed noise to match the gradient of the noisy input. This noise map is added to a random clean image to synthesize a new data pair, which is then fed back to the noise removal module to correct the noise removal process. These two modules cooperate to extract noise finely. After convergence, the noise removal module can remove noise without damaging other background details, so we use it as our final denoising network. Experiments show that the denoising performance of the proposed method is competitive with other supervised CNNs.

Comparison of Fenugreek Crop Evapotranspiration Measured by a Micro-lysimeter, Field Water Balance Method and Automatic Closed Canopy Chamber

The attempt was to design and develop an automatic closed canopy chamber (ACCC) having dimension of 1.2 m×1.2 m× 1.2 m for crop evapotranspiration measurement by using transparent acrylic sheet of 4 mm thickness. Between two small fans a temperature and relative humidity sensor was used to measure vapor density. The intermediate circuit was developed for making automation system in ACCC. The arduino based coding was developed as per desired logic operation. The top lead of chamber was automatically closed for 2 minutes when inside and outside temperature and relative humidity of ambient air were similar. During measurement mode of ACCC, the two fans were started automatically. After measurement mode, fans were automatically stopped and top lead was opened. The ACCC was calibrated by evaporating mass of water from water filled tray which was placed inside the automatic closed canopy chamber. The validation of the developed ACCC were made using micro-lysimeters (MLS) having size of 0.2 m × 0.2 m × 0.2 m by growing shallow rooted crop like fenugreek. The depth of irrigation was computed based on soil moisture content before irrigation and field capacity. The field testing of ACCC was made by placing chamber in plots of fenugreek crop. The irrigation was applied by drip irrigation as per crop water consumption. The sensor sensed and recorded the instantaneous temperature and relative humidity at 1 second interval for 24 hours. Two sample t-tests were done to compare the data pair of crop evapotranspiration obtained by the MLS inside the ACCC with that of outside the ACCC to ascertain whether there is any effect of the change in micro-climate for a short period of 2 minute on the crop growth physiological processes. Also, the data pair of crop evapotranspiration measured by the MLs, ACCC using the sensor data of temperature-relative humidity were compared and statistically analyzed through t-test. Similarly, the data pair of ETC measured by the FWB (Field Water Balance Method) and ACCC using the sensor data of temperature-relative humidity were also compared and statistically analyzed through t-test. The calibration factor of the ACCC was found as 1.666. The results revealed that there was no significant difference in the crop evapotranspiration measured by the MLs inside and outside ACCC. Also in case of validation and field testing of ACCC, there were no significant difference between the ETC measured by the ACCC, MLs and FWB at 95 percent confidence level. This implies that there are no effects of the change in micro-climate for a short period of 2 minutes in the chamber, on the plant physiological processes. The ETc rate of fenugreek increases as sun rises and reaches the peak after one to two hour from mid-day and then continuously decreases with time. During validation and field testing of ACCC, the fenugreek crop coefficients varied from 0.72 to 1.04 and 0.69 to 1.02 respectively. The developed ACCC is portable as well as more comfortable and cost effective compared to the lysimeter for the measurement of the actual crop evapotranspiration and the crop coefficient.

Developing a model to determine the number of vehicles lane changing on freeways by Brownian motion method

Lane change maneuvers are essential on car trips. Drivers change lanes to follow the desired route to reach their destination or improve their driving condition or level of service. To change lanes, a driver must consider several factors that affect safety. Due to the lack of appropriate data and consequently the lack of appropriate models to determine the number of lane-changes on the road (as an influential factor in accidents), this study attempts to collect proper data in a new way. Thus, the Qazvin-Karaj freeway was selected as the case study. After installing the imaging cameras and performing the image processing, SPSS and Expert Design statistical software were used to model development. The Brownian motion model was also used to construct the driver change lane model. The results showed that logarithmic model number 2 reported a better coefficient of determination than other models with a value of 0.472. Then models 3 and 9 were ranked with R 2 of 0.451 and 0.442, respectively. Also, the Expert Design model with R 2 (0.786) could have a better fit. The value of the response variable (Nch + 0.52)0.74 was obtained three-dimensionally against the changes of distance from the front vehicle (Df ) and distance from the rear vehicle (Db ). Variable values of distance from the front car and distance from the rear vehicle have more effective values on the number of lane changes than left and right distance values. The observed and Brownian data had a slight mean difference (0.018), and also, the standard deviation was so small. Also, the correlation in this data pair is 0.912, which is a suitable value and indicates a slight difference between the outputs of the Brownian model and the observations.

Utilization of Sentinel-1 satellite imagery data to support land subsidence analysis in DKI Jakarta, Indonesia

Land subsidence had been a significant problem in DKI Jakarta and Semarang, with at least 20 kilometres of roads affected. Repairing them will require at least US $ 1 million per kilometre. Land subsidence monitoring has been carried out using terrestrial methods (GPS and levelling), which are believed to have a high degree of accuracy. The high accuracy of the terrestrial method results in a lack of precision over a large area. On the other hand, remote sensing technology as a non-terrestrial method has developed to monitor land subsidence which can produce high precision over a large area. This study aimed to test the Sentinel-1 satellite data using the Differential Interferometric Synthetic Aperture Radar (DInSAR) method in monitoring land subsidence in DKI Jakarta. DInSAR is a method in Remote Sensing that utilizes radar sensors to analyze the phase differences of a SAR data pair that have different times of capture and have been catalogued to obtain displacement along the area of collection. The results showed that the North Jakarta area experienced the highest land subsidence in the entire Jakarta area. The annual average rate from 2017-2019 is 3.4 cm. The value of 3.4 cm is the average value of all samples in the North Jakarta area. The second area where high land subsidence is West Jakarta, where the maximum amount value of subsidence is 2.8 cm. The accuracy-test results with the MONAS test point showed that the difference between field data and DInSAR results was ± 6.5 cm. The results of this research indicate that the DInSAR method is quite capable of describing land subsidence in the DKI Jakarta area with a relatively good level of precision.

Is the surface salinity difference between the Atlantic and Indo–Pacific a signature of the Atlantic Meridional Overturning Circulation?

The high Atlantic surface salinity has sometimes been interpreted as a signature of the Atlantic Meridional Overturning Circulation and an associated salt advection feedback. Here, the role of oceanic and atmospheric processes for creating the surface salinity difference between the Atlantic and Indo–Pacific is examined using observations and a conceptual model. In each basin, zonally averaged data are represented in diagrams relating net evaporation (Ẽ) and surface salinity (S). The data-pair curves in the Ẽ – S plane share common features in both basins. However, the slopes of the curves are generally smaller in the Atlantic than in the Indo–Pacific, indicating a weaker sensitivity of the Atlantic surface salinity to net evaporation variations. To interpret these observations, a conceptual advective-diffusive model of the upper-ocean salinity is constructed. Notably, the Ẽ – S relations can be qualitatively reproduced with only meridional diffusive salt transport. In this limit, the inter-basin difference in salinity is caused by the spatial structure of net evaporation, which in the Indo–Pacific oceans contains lower meridional wavenumbers that are weakly damped by the diffusive transport. The observed Atlantic Ẽ – S relationship at the surface reveals no clear influence of northward advection associated with the meridional overturning circulation; however a signature of northward advection emerges in the relationship when the salinity is vertically averaged over the upper kilometer. The results indicate that the zonal-mean near-surface salinity is shaped primarily by the spatial pattern of net evaporation and the diffusive meridional salt transport due to wind-driven gyres and mesoscale ocean eddies, rather than by salt advection within the meridional overturning circulation.

Indirect Validation of Ocean Remote Sensing Data via Numerical Model: An Example of Wave Heights from Altimeter

Using numerical model outputs as a bridge, an indirect validation method for remote sensing data was developed to increase the number of effective collocations between remote sensing data to be validated and reference data. The underlying idea for this method is that the local spatial-temporal variability of specific parameters provided by numerical models can compensate for the representativeness error induced by differences of spatial-temporal locations of the collocated data pair. Using this method, the spatial-temporal window for collocation can be enlarged for a given error tolerance. To test the effectiveness of this indirect validation approach, significant wave height (SWH) data from Envisat were indirectly compared against buoy and Jason-2 SWHs, using the SWH gradient information from a numerical wave hindcast as a bridge. The results indicated that this simple indirect validation method is superior to “direct” validation.

Heterogeneous Transfer Learning with Weighted Instance-Correspondence Data

Instance-correspondence (IC) data are potent resources for heterogeneous transfer learning (HeTL) due to the capability of bridging the source and the target domains at the instance-level. To this end, people tend to use machine-generated IC data, because manually establishing IC data is expensive and primitive. However, existing IC data machine generators are not perfect and always produce the data that are not of high quality, thus hampering the performance of domain adaption. In this paper, instead of improving the IC data generator, which might not be an optimal way, we accept the fact that data quality variation does exist but find a better way to use the data. Specifically, we propose a novel heterogeneous transfer learning method named Transfer Learning with Weighted Correspondence (TLWC), which utilizes IC data to adapt the source domain to the target domain. Rather than treating IC data equally, TLWC can assign solid weights to each IC data pair depending on the quality of the data. We conduct extensive experiments on HeTL datasets and the state-of-the-art results verify the effectiveness of TLWC.

THUNDERSTORM WEATHER ANALYSIS BASED ON XGBOOST ALGORITHM

Using the GPS data service platform of China seismological bureau to get the ZTD separated ZWD data pair and the content in the air, and by detecting the O3 value in the air is an effective method to analyze and study the thunderstorm weather.This paper collected the four foundations of the beibu gulf region GPS station in 10 days in August 2019 data, through ZWD numerical and O3 values after consolidation, the classification of the training and testing, in XGboost algorithm, manual adjustment method is compared with grid search method, and the results show that the model of manual adjustment method is superior to grid search model and the default model in accuracy and AUC value.