We provide an introduction of the functioning, implementation, and challenges of convolutional neural networks (CNNs) to classify visual information in social sciences. This tool can help scholars to make more efficient the tedious task of classifying images and extracting information from them. We illustrate the implementation and impact of this methodology by coding handwritten information from vote tallies. Our paper not only demonstrates the contributions of CNNs to both scholars and policy practitioners, but also presents the practical challenges and limitations of the method, providing advice on how to deal with these issues.

Annual resolution sediment layers, known as varves, can provide continuous and high-resolution chronologies of sedimentary sequences. In addition, varve counting is not burdened with the high laboratory costs of geochronological analyses. Despite a more than 100-year history of use, many existing varve counting techniques are time consuming and difficult to reproduce. We present countMYvarves, a varve counting toolbox which uses sliding-window autocorrelation to count the number of repeated patterns in core scans or outcrop photos. The toolbox is used to build an annually-resolved record of sedimentation rates, which are depth-integrated to provide ages. We validate the model with repeated manual counts of a high sedimentation rate lake with biogenic varves (Herd Lake, USA) and a low sedimentation rate glacial lake (Lago Argentino, Argentina). In both cases, countMYvarves is consistent with manual counts and provides additional sedimentation rate data. The toolbox performs multiple simultaneous varve counts, enabling uncertainty to be quantified and propagated into the resulting age-depth model. The toolbox also includes modules to automatically exclude non-varved portions of sediment and interpolate over missing or disrupted sediment. CountMYvarves is open source, runs through a graphical user interface, and is available online for download for use on Windows, macOS or Linux at https://doi.org/10.5281/zenodo.4031811.

Symmetry is omnipresent in nature and we encounter symmetry routinely in our everyday life. It is also common on the microscopic level, where symmetry is often key to the proper function of core biological processes. The human brain is exquisitely well suited to recognize such symmetrical features with ease. In contrast, computational recognition of such patterns in images is still surprisingly challenging. In this paper we describe a mathematical approach to identifying smaller local symmetrical structures within larger images. Our algorithm attributes a local symmetry score to each image pixel, which subsequently allows the identification of the symmetrical centers of an object. Though there are already many methods available to detect symmetry in images, to the best of our knowledge, our algorithm is the first that is easily applicable in ImageJ/FIJI. We have created an interactive plugin in FIJI that allows the detection and thresholding of local symmetry values. The plugin combines the different reflection symmetry axis of a square to get a good coverage of reflection symmetry in all directions. To demonstrate the plugins potential, we analyzed images of bacterial chemoreceptor arrays and intracellular vesicle trafficking events, which are two prominent examples of biological systems with symmetrical patterns.

Deep learning has pushed the scope of digital pathology beyond simple digitization and telemedicine. The incorporation of these algorithms in routine workflow is on the horizon and maybe a disruptive technology, reducing processing time, and increasing detection of anomalies. While the newest computational methods enjoy much of the press, incorporating deep learning into standard laboratory workflow requires many more steps than simply training and testing a model. Image analysis using deep learning methods often requires substantial pre- and post-processing order to improve interpretation and prediction. Similar to any data processing pipeline, images must be prepared for modeling and the resultant predictions need further processing for interpretation. Examples include artifact detection, color normalization, image subsampling or tiling, removal of errant predictions, etc. Once processed, predictions are complicated by image file size – typically several gigabytes when unpacked. This forces images to be tiled, meaning that a series of subsamples from the whole-slide image (WSI) are used in modeling. Herein, we review many of these methods as they pertain to the analysis of biopsy slides and discuss the multitude of unique issues that are part of the analysis of very large images.

Time-resolved imaging of molecules and materials made of light elements is an emerging field of transmission electron microscopy (TEM), and the recent development of direct electron detection cameras, capable of taking as many as 1,600 fps, has potentially broadened the scope of the time-resolved TEM imaging in chemistry and nanotechnology. However, such a high frame rate reduces electron dose per frame, lowers the signal-to-noise ratio (SNR), and renders the molecular images practically invisible. Here, we examined image noise reduction to take the best advantage of fast cameras and concluded that the Chambolle total variation denoising algorithm is the method of choice, as illustrated for imaging of a molecule in the 1D hollow space of a carbon nanotube with ~1 ms time resolution. Through the systematic comparison of the performance of multiple denoising algorithms, we found that the Chambolle algorithm improves the SNR by more than an order of magnitude when applied to TEM images taken at a low electron dose as required for imaging at around 1,000 fps. Open-source code and a standalone application to apply Chambolle denoising to TEM images and video frames are available for download.

Multicomponent polymer systems are of interest in organic photovoltaic and drug delivery applications, among others where diverse morphologies influence performance. An improved understanding of morphology classification, driven by composition-informed prediction tools, will aid polymer engineering practice. We use a modified Cahn–Hilliard model to simulate polymer precipitation. Such physics-based models require high-performance computations that prevent rapid prototyping and iteration in engineering settings. To reduce the required computational costs, we apply machine learning (ML) techniques for clustering and consequent prediction of the simulated polymer-blend images in conjunction with simulations. Integrating ML and simulations in such a manner reduces the number of simulations needed to map out the morphology of polymer blends as a function of input parameters and also generates a data set which can be used by others to this end. We explore dimensionality reduction, via principal component analysis and autoencoder techniques, and analyze the resulting morphology clusters. Supervised ML using Gaussian process classification was subsequently used to predict morphology clusters according to species molar fraction and interaction parameter inputs. Manual pattern clustering yielded the best results, but ML techniques were able to predict the morphology of polymer blends with ≥90% accuracy.

This research attempts to systematically establish shape descriptor states through elliptic Fourier analysis (EFA) using pili (Canarium ovatum Engl.) kernel as a model. Kernel images of 53 pili accessions from the National Plant Genetic Resources Laboratory (NPGRL), University of the Philippines Los Baños were acquired using VideometerLab 3. Shape features, such as roundness, compactness and elongation, were extracted from the images. Shapes outlines were characterized using elliptic Fourier coefficients calculated from SHAPE version 1.3 software. Principal component analysis and cluster analysis were used to elucidate clusters representing the shape descriptor states. The first principal component accounts for the variation in length to width ratio; whereas, the second and third principal components explain the variation in the location of the widest portion and the truncation of the apex and base of the kernel, respectively. Cluster analysis separated the different accessions into six distinct clusters at 0.04 Euclidian distance. Six descriptor states, narrowly elliptic, elliptic, widely elliptic, ovate, obovate and lance-ovate, were characterized from the shape outlines and visualized through R's shape on r package. The discrimination between clusters was validated through MANOVA and LDA with 95% correct classification. The Fourier coefficients were also able to represent the variation observed from the physical properties of shape. The method may be used in establishing shape descriptors of all plant parts of all crop species.

Economic pressures continue to mount on modern-day livestock farmers, forcing them to increase herds sizes in order to be commercially viable. The natural consequence of this is to drive the farmer and the animal further apart. However, closer attention to the animal not only positively impacts animal welfare and health but can also increase the capacity of the farmer to achieve a more sustainable production. State-of-the-art precision livestock farming (PLF) technology is one such means of bringing the animals closer to the farmer in the facing of expanding systems. Contrary to some current opinions, it can offer an alternative philosophy to ‘farming by numbers’. This review addresses the key technology-oriented approaches to monitor animals and demonstrates how image and sound analyses can be used to build ‘digital representations’ of animals by giving an overview of some of the core concepts of PLF tool development and value discovery during PLF implementation. The key to developing such a representation is by measuring important behaviours and events in the livestock buildings. The application of image and sound can realise more advanced applications and has enormous potential in the industry. In the end, the importance lies in the accuracy of the developed PLF applications in the commercial farming system as this will also make the farmer embrace the technological development and ensure progress within the PLF field in favour of the livestock animals and their well-being.

Silver nanowire (AgNW) diameters are typically characterized by manual measurement from high magnification electron microscope images. Measurement is monotonous and has potential ergonomic hazards. Because of this, statistics regarding wire diameter distribution can be poor, costly, and low-throughput. In addition, manual measurements are of unknown uncertainty and operator bias. In this paper we report an improved microscopy method for diameter and yield measurement of nanowires in terms of speed/automation and reduction of analyst variability. Each step in the process to generate these measurements was analyzed and optimized: microscope imaging conditions, sample preparation for imaging, image acquisition, image analysis, and data processing. With the resulting method, average diameter differences between samples of just a few nanometers can be confidently and statistically distinguished, allowing the identification of subtle incremental improvements in reactor processing conditions, and insight into nucleation and growth kinetics of AgNWs.

Swedish nursing homes’ use of Instagram has increased vastly in the past few years. Instagram is understood as a means to manage the image they wish to mediate to the public. This article examines what is displayed in the nursing homes’ Instagram accounts, and what kind of reality is thereby constructed. The data consist of 338 Instagram images from four nursing homes’ Instagram accounts. It is found that nursing home life is primarily depicted on Instagram as active, sociable and fun, with informal, friendly relations between staff and residents, and residents able to continue to live as before, if not better, and to interact with surrounding society. Frailty, boredom, loneliness and death were absent from the data, as were mundane care activities. The article concludes that the presentations in the Instagram accounts challenge the traditional idea of nursing homes as total institutions, and the decline and loss associated with living in such institutions; however, there is a risk that these idyllic presentations conceal the inherent problems of nursing home life.

In the present study, precise, animal-based biometric data on the space needed for the body dimensions of individual pigs (static space) were collected. Per batch, two groups of eight piglets each were formed after weaning (35 days old). Using three-dimensional cameras that recorded a piglets’ pen from above and newly developed software, the static space of individuals was determined over 6 weeks. The area covered by an individual increased almost linearly with increasing body weight (R2 = 0.97). At the end of rearing (25 kg body weight), an individual covered 1704 cm2 in standing position, 1687 cm2 in sitting posture and 1798 cm2 in a recumbent position. According to the allometric equation: Space = k × body weight0.667, k values for the static space in standing position (k = 0.021), in recumbent position in general (k = 0.022) and in lateral recumbent posture (k = 0.027) were calculated. Compared with spatial requirements in different countries, the results of static space obtained in the present study revealed that pigs weighing 25 kg are provided with 0.09–0.18 m2 free space per pig which is not covered by the pig's body. This free space can be used as dynamic space needed for body movements or social interactions. The present study was not intended to enhance space recommendations in pig farming, but to demonstrate the amount of free space in a pigs’ pen. It was shown that innovative technologies based on image analysis offer completely new possibilities to assess spatial requirements for pigs.

Coralligenous bioconstructions are among the most important Mediterranean habitats for biodiversity maintenance. However some characteristic and sensitive organisms, such as the fan corals, are considered endangered in the international community; indeed, they may be severely damaged by fishing activities causing mechanical damage and increasing sedimentation rate. ROV (Remotely Operated Vehicle) investigations were carried out in order to characterize different morphological types of coralligenous habitat (rim, bank and shoal) located in the southern Bay of Naples (Italy), and to assess the presence of lost fishing gears and their impact on these benthic communities. A rapid classification of different fishing pressures and impacts was obtained through the development of new, representative and synthetic categories. Image analysis revealed the presence of rich and diversified communities, characterized by several fan coral colonies. However, fishing activity dramatically affects these coralligenous habitats, entangling and covering benthic assemblages and leading to necrosis and to parasitic epibionts growth especially on branched organisms. Monitoring programmes may provide a detailed assessment of coralligenous habitats characterization, distribution and health status. An accurate evaluation of fishing pressure and impact may be considered a useful tool to improve sustainable management of these valuable habitats.

An increasing number of farm machines nowadays implement precision agriculture technologies. Most of these operate through proximal sensing using optical sensors (i.e. NIR or Vis-NIR). Imaging techniques in this context have received minor consideration due to the complex analysis of the data but on the other side offer great flexibility. This study reports a preliminary pilot imaging multi-sensor retrofit system to be applied independently on a wide range of agricultural machines and able to test different monitoring or control image-based applications for precision agriculture. The process, based on RGB image, was tested for in-field discrimination of weeds in lettuce and broccoli crops. It works by discriminating and extracting single plants from the soil and weeds. However, to be truly implementable, the experimental code should be optimized in order to shorten the time needed for acquisition and processing.

Anisotropy of magnetic susceptibility (AMS) is a frequently applied method in sedimentology, especially in the determination of the orientation of transport processes. We present an analysis of magnetic fabric (MF) studies on loess. New aspects of fabric development reveal: i) The deposition of the aeolian sediments was controlled by gravity, low-energy transport and local geomorphology, hence no clarified wind direction can be defined. ii) The influence of phyllosilicates is also significant among the magnetic components. iii) While the primary MF is relatively well-defined, the secondary MF is influenced by several processes. The analysis of stereoplots combined with the q—β diagram and photostatistics showed encouraging results during the characterization of various secondary MF such as redeposited MF and pedogenic fabric. iv) Changes in processes from aeolian to water-lain deposition and the increasing transportation energy were reflected by the connection between AMS and observed micro-scale sedimentary features. v) A relationship was obvious between the degree of pedogenesis and the transformation of sedimentary MF into a vertical MF typical for paleosols. vi) The significant role of very fine grained magnetite on the formation of inverse MF could not be excluded.

Different quartz types from several localities in the Czech Republic and Sweden were examined by polarizing microscopy combined with cathodoluminescence (CL) microscopy, spectroscopy, and petrographic image analysis, and tested by use of an accelerated mortar bar test (following ASTM C1260). The highest alkali–silica reaction potential was indicated by very fine-grained chert, containing significant amounts of fine-grained to cryptocrystalline matrix. The chert exhibited a dark red CL emission band at ~640 nm with a low intensity. Fine-grained orthoquartzites, as well as fine-grained metamorphic vein quartz, separated from phyllite exhibited medium expansion values. The orthoquartzites showed various CL of quartz grains, from blue through violet, red, and brown. Two CL spectral bands at ~450 and ~630 nm, with various intensities, were detected. The quartz from phyllite displayed an inhomogeneous dark red CL with two CL spectral bands of low intensities at ~460 and ~640 nm. The massive coarse-grained pegmatite quartz from pegmatite was assessed to be nonreactive and displayed a typical short-lived blue CL (~480 nm). The higher reactivity of the fine-grained hydrothermal quartz may be connected with high concentrations of defect centers, and probably with amorphized micro-regions in the quartz, respectively; indicated by a yellow CL emission (~570 nm).

This study introduces a passive autofocus method based on image analysis calculating the Bayes spectral entropy (BSE). The method is applied to optical microscopy and together with the specific construction of the opto-mechanical unit, it allows the analysis of large samples with complicated surfaces without subsampling. This paper will provide a short overview of the relevant theory of calculating the normalized discrete cosine transform when analyzing obtained images, in order to find the BSE measure. Furthermore, it will be shown that the BSE measure is a strong indicator, helping to determine the focal position of the optical microscope. To demonstrate the strength and robustness of the microscope system, tests have been performed using a 1951 USAF test pattern resolution chart determining the in focus position of the microscope. Finally, this method and the optical microscope system is applied to analyze an optical grating (100 lines/mm) demonstrating the detection of the focal position. The paper concludes with an outlook of potential applications of the presented system within quality control and surface analysis.

In biology, hemocytometers such as Malassez slides are widely used and are effective tools for counting cells manually. In a previous work, a robust algorithm was developed for grid extraction in Malassez slide images. This algorithm was evaluated on a set of 135 images and grids were accurately detected in most cases, but there remained failures for the most difficult images. In this work, we present an optimization of this algorithm that allows for 100% grid detection and a 25% improvement in grid positioning accuracy. These improvements make the algorithm fully reliable for grid detection. This optimization also allows complete erasing of the grid without altering the cells, which eases their segmentation.

Mechanical properties of the arterial wall depend largely on orientation and density of collagen fiber bundles. Several methods have been developed for observation of collagen orientation and density; the most frequently applied collagen-specific manual approach is based on polarized light (PL). However, it is very time consuming and the results are operator dependent. We have proposed a new automated method for evaluation of collagen fiber direction from two-dimensional polarized light microscopy images (2D PLM). The algorithm has been verified against artificial images and validated against manual measurements. Finally the collagen content has been estimated. The proposed algorithm was capable of estimating orientation of some 35 k points in 15 min when applied to aortic tissue and over 500 k points in 35 min for Achilles tendon. The average angular disagreement between each operator and the algorithm was −9.3±8.6° and −3.8±8.6° in the case of aortic tissue and −1.6±6.4° and 2.6±7.8° for Achilles tendon. Estimated mean collagen content was 30.3±5.8% and 94.3±2.7% for aortic media and Achilles tendon, respectively. The proposed automated approach is operator independent and several orders faster than manual measurements and therefore has the potential to replace manual measurements of collagen orientation via PLM.

Algebraic Multigrid (AMG) methods were developed originally for numerically solving Partial Differential Equations (PDE), not necessarily on structured grids. In the last two decades solvers inspired by the AMG approach, were developed for non PDE problems, including data and image analysis problems, such as clustering, segmentation, quantization and others. These solvers share a common principle in that there is a crosstalk between fine and coarse representations of the problems, with flow of information in both directions, fine-to-coarse and coarse-to-fine. This paper surveys some of these problems and the AMG-inspired algorithms for their solution.