Estimating sliding drop width via side-view features using recurrent neural networks.

High speed side-view videos of sliding drops enable researchers to investigate drop dynamics and surface properties. However, understanding the physics of sliding requires knowledge of the drop width. A front-view perspective of the drop is necessary. In particular, the drop's width is a crucial parameter owing to its association with the friction force. Incorporating extra cameras or mirrors to monitor changes in the width of drops from a front-view perspective is cumbersome and limits the viewing area. This limitation impedes a comprehensive analysis of sliding drops, especially when they interact with surface defects. Our study explores the use of various regression and multivariate sequence analysis (MSA) models to estimate the drop width at a solid surface solely from side-view videos. This approach eliminates the need to incorporate additional equipment into the experimental setup. In addition, it ensures an unlimited viewing area of sliding drops. The Long Short Term Memory (LSTM) model with a 20 sliding window size has the best performance with the lowest root mean square error (RMSE) of 67 µm. Within the spectrum of drop widths in our dataset, ranging from 1.6 to 4.4 mm, this RMSE indicates that we can predict the width of sliding drops with an error of 2.4%. Furthermore, the applied LSTM model provides a drop width across the whole sliding length of 5 cm, previously unattainable.

Control of spontaneous charging of sliding water drops by plasma-surface treatment.

Slide electrification is the spontaneous separation of electric charges at the rear of water drops sliding over solid surfaces. This study delves into how surfaces treated with a low-pressure plasma impact water slide electrification. Ar, O2, and N2 plasma treatment reduced the drop charge and contact angles on glass, quartz, and SU-8 coated with 1H,1H,2H,2H-perfluoroctyltrichlorosilane (PFOTS), and polystyrene. Conversely, 64% higher drop charge was achieved using electrode-facing treatment in plasma chamber. Based on the zeta potential, Kelvin potential, and XPS measurements, the plasma effects were attributed to alterations of the topmost layer's chemistry, such as oxidation and etching, and superficially charge deposition. The surface top layer charges were less negative after electrode-facing and more negative after bulk plasma treatment. As a result, the zeta potential was less negative after electrode-facing and more negative after bulk plasma treatment. Although the fluorinated layer was applied after plasma activation, we observed a discernible impact of plasma-glass treatment on drop charging. Plasma surface modification offers a means to adjust drop charges: electrode-facing treatment of the fluorinated layer leads to an enhanced drop charge, while plasma treatment on the substrate prior to fluorination diminishes drop charges, all without affecting contact angles or surface roughness.

Deep Learning to Analyze Sliding Drops.

State-of-the-art contact angle measurements usually involve image analysis of sessile drops. The drops are symmetric and images can be taken at high resolution. The analysis of videos of drops sliding down a tilted plate is hampered due to the low resolution of the cutout area where the drop is visible. The challenge is to analyze all video images automatically, while the drops are not symmetric anymore and contact angles change while sliding down the tilted plate. To increase the accuracy of contact angles, we present a 4-segment super-resolution optimized-fitting (4S-SROF) method. We developed a deep learning-based super-resolution model with an upscale ratio of 3; i.e., the trained model is able to enlarge drop images 9 times accurately (PSNR = 36.39). In addition, a systematic experiment using synthetic images was conducted to determine the best parameters for polynomial fitting of contact angles. Our method improved the accuracy by 21% for contact angles lower than 90° and by 33% for contact angles higher than 90°.

Scanning Drop Friction Force Microscopy.

Wetting imperfections are omnipresent on surfaces. They cause contact angle hysteresis and determine the wetting dynamics. Still, existing techniques (e.g., contact angle goniometry) are not sufficient to localize inhomogeneities and image wetting variations. We overcome these limitations through scanning drop friction force microscopy (sDoFFI). In sDoFFI, a 15 µL drop of Milli-Q water is raster-scanned over a surface. The friction force (lateral adhesion force) acting on the moving contact line is plotted against the drop position. Using sDoFFI, we obtained 2D wetting maps of the samples having sizes in the order of several square centimeters. We mapped areas with distinct wetting properties such as those present on a natural surface (e.g., a rose petal), a technically relevant superhydrophobic surface (e.g., Glaco paint), and an in-house prepared model of inhomogeneous surfaces featuring defined areas with low and high contact angle hysteresis. sDoFFI detects features that are smaller than 0.5 mm in size. Furthermore, we quantified the sliding behavior of drops across the boundary separating areas with different contact angles on the model sample. The sliding of a drop across this transition line follows a characteristic stick-slip motion. We use the variation in force signals, advancing and receding contact line velocities, and advancing and receding contact angles to identify zones of stick and slip. When scanning the drop from low to high contact angle hysteresis, the drop undergoes a stick-slip-stick-slip motion at the interline. Sliding from high to low contact angle hysteresis is characterized by the slip-stick-slip motion. The sDoFFI is a new tool for 2D characterization of wetting properties, which is applicable to laboratory-based samples but also characterizes biological and commercial surfaces.

Persian sentiment analysis of an online store independent of pre-processing using convolutional neural network with fastText embeddings.

Sentiment analysis plays a key role in companies, especially stores, and increasing the accuracy in determining customers' opinions about products assists to maintain their competitive conditions. We intend to analyze the users' opinions on the website of the most immense online store in Iran; Digikala. However, the Persian language is unstructured which makes the pre-processing stage very difficult and it is the main problem of sentiment analysis in Persian. What exacerbates this problem is the lack of available libraries for Persian pre-processing, while most libraries focus on English. To tackle this, approximately 3 million reviews were gathered in Persian from the Digikala website using web-mining techniques, and the fastText method was used to create a word embedding. It was assumed that this would dramatically cut down on the need for text pre-processing through the skip-gram method considering the position of the words in the sentence and the words' relations to each other. Another word embedding has been created using the TF-IDF in parallel with fastText to compare their performance. In addition, the results of the Convolutional Neural Network (CNN), BiLSTM, Logistic Regression, and Naïve Bayes models have been compared. As a significant result, we obtained 0.996 AUC and 0.956 F-score using fastText and CNN. In this article, not only has it been demonstrated to what extent it is possible to be independent of pre-processing but also the accuracy obtained is better than other researches done in Persian. Avoiding complex text preprocessing is also important for other languages since most text preprocessing algorithms have been developed for English and cannot be used for other languages. The created word embedding due to its high accuracy and independence of pre-processing has other applications in Persian besides sentiment analysis.