Enhancing agricultural automation through weather invariant soil parameter prediction using machine learning.

Soil parameters are crucial aspects in increasing agricultural production. Even though Bangladesh is heavily dependent on agriculture, little research has been done regarding its automation. And a vital aspect of agricultural automation is predicting soil parameters. Generally, sensors relating to soil parameters are quite expensive and are often done in a controlled environment such as a greenhouse. However, a large scale implementation of such expensive sensors is not very feasible. This work tries to find an inexpensive solution towards predicting soil parameters such as soil moisture and temperature, both of which are crucial to the growth of crops. We focus on finding a robust relation between the above mentioned soil parameters with the nearby weather parameters such as humidity and temperature, irrespective of the weather. We apply different machine learning models like multilayer perceptron (MLP), random forest, etc. to predict the soil parameters, given the humidity and temperature of the surrounding environment. For all the experiments we have used a custom made dataset, which contains around 9000 datapoints of soil moisture & temperature, ambient humidity & temperature. The data has been collected in an uncontrolled agriculture bed via inexpensive sensors. Our results show that XGBoost regressor achieves the best results with an R2 score of 0.93 and 0.99 for soil moisture and soil temperature data respectively. This suggests very high correlation between the weather parameters and soil parameters. The model also portrayed a very low root mean squared error and mean absolute error of 0.037 & 0.015 for soil moisture and 0.001 & 0.0008 for soil temperature. Our results show that it is indeed possible to find the soil parameters from the corresponding weather, which will have great impact on mass agricultural automation. The dataset has been made publicly available at https://github.com/Nadimulhaque0403/Soil_parameter_prediction_dataset.

COVIDXception-Net: A Bayesian Optimization-Based Deep Learning Approach to Diagnose COVID-19 from X-Ray Images.

COVID-19 is spreading around the world like wildfire. Chest X-rays are used as one of the primary tools for diagnosing COVID-19. However, about two-thirds of the world population do not have access to sufficient radiological services. In this work, we propose a deep learning-driven automated system, COVIDXception-Net, for diagnosing COVID-19 from chest X-rays. A primary challenge in any data-driven COVID-19 detection is the scarcity of COVID-19 data, which heavily deteriorates a deep learning model's performance. To address this issue, we incorporate a weighted-loss function that ensures the COVID-19 cases are given more importance during the training process. We also propose using Bayesian Optimization to find the best architecture for detecting COVID-19. Extensive experimentation on four publicly available COVID-19 datasets shows that our proposed model achieves an accuracy of 0.94, precision 0.95, recall 0.94, specificity 0.997, F1-score 0.94, and Matthews correlation coefficient 0.992 outperforming three widely used architectures-VGG16, MobileNetV2, and InceptionV3. It also surpasses the performance of several state-of-the-art COVID-19 detection methods. We also performed two ablation studies that show our model's accuracy degrades from 0.994 to 0.950 when a random search is used and to 0.983 when a regular loss function is employed instead of the Bayesian and weighted loss, respectively.

Solar powered nanofiltration for drinking water production from fluoride-containing groundwater - A pilot study towards developing a sustainable and low-cost treatment plant.

The following paper summarizes the findings of a pilot study to develop a simple, low-cost, holistic water concept on fluoride removal from groundwater in rural communities of Tanzania; an ideal representative community for other areas in the world with similar problems. A small photovoltaic powered nanofiltration (NF) pilot plant was installed at a vocational training center in Boma Ng´ombe in northern Tanzania. The groundwater in this region is contaminated with fluoride at very high concentrations of up to 60 mg/L. The pilot plant was equipped with a single membrane module containing a spiral wound 4040 membrane NF90 of Dow Water & Process Solutions and was successfully operated over a nine-month period. The membrane removed more than 98% of fluoride. In fact, the fluoride concentration in the permeate was always less than 1 mg/L, which is in agreement with the WHO recommended standard (1.5 mg/L). Permeate was also used as weekly flush medium, so no chemical cleaning was required. Aside from permeate (drinking water) concentrate was also used for washing and flushing the toilets. In conclusion, the use of solar PV power (2.25 KWP) for approximately 2.5 h per day allowed producing about 240 L/h of permeate on average. Therefore, the sustainability of the process and suitability for the Tanzanian communities was proved.

Novel low-fouling membranes from lab to pilot application in textile wastewater treatment.

A novel antifouling coating based on the polymerization of a polymerisable bicontinuous microemulsion (PBM) was developed and applied for commercially available membranes for textile wastewater treatment. PBM coating was produced by polymerizing, on a polyethersulfone (PES) membrane, a bicontinuous microemulsion, realized by finely tuning its properties in terms of chemical composition and polymerization temperature. In particular, the PBM was prepared by using, as the surfactant component, inexpensive and commercially available dodecyltrimethylammonium bromide (DTAB). The coating exhibited a more hydrophilic and a smoother surface in comparison to uncoated PES surface, making the produced PBM membranes more resistant and less prone to be affected by fouling. The anti-fouling potential of PBM membranes was assessed by using humic acid (HA) as a model foulant, evaluating the water permeability decrease as an indicator of the fouling propensity of the membranes. PBM membrane performances in terms of dye rejection, when applied for model textile wastewater treatment, were also evaluated and compared to PES commercial ones. The PBM membranes were finally successfully scaled-up (total membrane area 0.33 m2) and applied in a pilot membrane bioreactor (MBR) unit for the treatment of real textile wastewater.