Visual Observation Research Articles

New geopolymer preparation methods and its application by recycle of double-waste: ground granulated blast furnace slag and fly ash

ABSTRACT This article mainly studies new geopolymer preparation methods by recycle of double waste:ground granulated blast furnace slag (GGBFS) and fly ash as geopolymer precursor, through different mass ratio of GGBFS and fly ash, to prepare geopolymers by alkali activation, the samples is cured at ambient temperature and the mechanical test is carried out together with the analysis of XRD, SEM and EDS to research the mechanism of strengthening and geopolymerisation. After analysis we find that the sample prepared by 30% mass ratio of GGBFS have good initial strength and short setting time, as well as best mechanical performance, which is attributed to moderate Ca2+ introduction accelerates the speed of geopolymerisation and the reactants of N-A-S-H gel, C-A-S-H gel and hydrated C-S-H help to bridge the gaps between the different hydrated phases, unreacted particles and matrix, result in a homogeneous and compacted geopolymeric sample. When GGBFS mass ratio above 50%, the mechanical of samples is deteriorated due to too much hydrated calcium containing compounds are formed and covered on the unreacted GGBFS particles surface to block further geopolymerisation. The acid attack test and water resistance test are carried out to analysis the durability between 30% GGBFS introduced geopolymer and ordinary Portland cement (OPC). In additional, a field application in case of 30% GGBFS introduced geopolymer grouting repair material indicated desirable construction based on visual observation and ground-penetration radar scanning.

Post-occupancy evaluation for tactical urbanism interventions through placemaking model: two case studies from the “Piazze Aperte” program in Milan

ABSTRACT This research examines whether tactical urbanism is capable of generating successful public places based on key placemaking principles. Drawing from established evaluation models, a new model is designed and adapted to the tactical urbanism interventions. The proposed model consists of six principal qualities and 22 variables that are evaluated through a mixed-methods approach. To test its effectiveness, two of the public spaces of the “Piazze Aperte” program in Milan were analyzed, allowing for a comprehensive evaluation of their post-occupancy in providing an inclusive, social, and functional place. The analysis is built on several tools including visual observations, in-depth interviews, and GIS spatial analysis mapping. The findings include the identification of best practices for tactical urbanism as well as limitations and areas for improvement. Lastly, a reflection on the post-occupancy status of the case studies towards the inclusion of community and stakeholders in future planning processes of the “Piazze Aperte” program.

Tunable, reagent‐loaded polyurethane nanocapsules cleavable by NIR light

AbstractDual response polyurethane nanocapsules consisting of a hydrophilic core are synthesized via interfacial polyaddition of the diisocyanate monomers and different diols (VA‐060 and glycols) in inverse miniemulsion process. The presence of the water‐soluble NIR dye in the core and azo‐bonds in the polymer shell allows the selective release of encapsulated material triggered by temperature or NIR light. The capsules are characterized by transmission electron microscopy (TEM) and dynamic light scattering (DLS). The capsule degradation under external stimuli, like temperature and NIR light, is confirmed microscopically by TEM. Macroscopic evidence of the capsule cleavage was achieved by the incorporation of the chemical into the capsule core, and subsequent treatment of the capsules with NIR laser in the presence of the suitable reagent outside the capsules. A color‐forming chemical reaction occurred after the shell opening. The reactions were easily detected by visual observation of a color change and by UV–Vis spectroscopy.

A green zwitterionic bi-continuous membranes prepared by dual-bath procedure for mitigating biofouling during bacterial removal from water

This work explores the possibility of forming green microfiltration antifouling membranes based on polyvinylidene fluoride (PVDF) and a zwitterionic copolymer derivative of sulfobetaine methacrylate (SBMA), utilizing gamma-valerolactone (GVL) as the solvent and a mixture of ethanol and water as non-solvents. Efforts were initially focused on adjusting the immersion time of the matrix polymer in alcohol alone to obtain a porous structure, and 15 s were sufficient to achieve a symmetric, open porous morphology. Next, the polymer/copolymer/solvent blend was characterized. While it did not show the characteristics of a fully miscible system, DLS and visual observations revealed the obtaining of a homogeneous (relatively narrow particle size) and transparent to translucent system. Thus, membranes could be cast, and were shown to be stable. The addition of the copolymer maintained the physical characteristics of microfiltration membranes, but an FT-IR analysis revealed its effect on the dominant crystalline polymorph (switching from β to α). The membrane wettability was enhanced in an important extent with a dynamic water contact angle (WCA) gradually falling to < 40° (while the virgin membrane's WCA remained about 138° all test long), and a hydration capacity reaching about 530 mg/cm3 against <10 mg/cm3 for the virgin membrane. It reflected on the antifouling abilities of the membranes as tested using bovine serum albumin, fibrinogen, Escherichia coli, and Stenotrophomonas maltophilia, with relative decreases in adsorption during static tests measured to be 95 %, 70 %, 91 % and 98 %, respectively. Finally, compared to the virgin membrane, the modified membrane featured improved flux recovery (32 % vs. 11 %), significantly higher rejection (98.9 % vs. 89.1 %) and lower irreversible fouling tendency (68 % vs. 89 %) during cyclic water/bacterial solution filtration in dead-end mode, making it potentially suitable for addressing bacterial pollution in wastewater.

Integrating vision and somatosensation does not improve the accuracy and response time when estimating area and perimeter of rectangles in primary school

BackgroundProblem-solving and learning in mathematics involves sensory perception and processing. Multisensory integration may contribute by enhancing sensory estimates. This study aims to assess if combining visual and somatosensory information improves elementary students' perimeter and area estimates. Methods87 4th graders compared rectangles with respect to area or perimeter either solely using visual observation or additionally with somatosensory information. Three experiments targeted different task aspects. Statistical analyses tested success rates and response times. ResultsContrary to expectations, adding somatosensory information did not boost success rates for area and perimeter comparison. Response time even increased with adding somatosensory information. Children's difficulty in accurately tracing figures negatively impacted the success rate of area comparisons. DiscussionResults suggest visual observation alone suffices for accurately estimating and comparing area and perimeter of rectangles in 4th graders. ImplicationsCareful deliberation on the inclusion of somatosensory information in mathematical tasks concerning perimeter and area estimations of rectangles is recommended.

Large-scale assessment of characteristic plant species on Eurasian saline and alkaline soda ecosystems

From an ecological standpoint, saline and alkaline habitats are considered as extreme environments for vascular plants because salinity and alkalinity serve as primary structuring factors that significantly influence the species richness of vegetation cover. The non-alkaline salinity stress on plants primarily results from interactions of NaCl, Na2SO4, and other neutral salts, whereas alkalinity stress is caused by NaHCO3 and Na2CO3, accompanied by an increase in pH. Some halophytes and salt-tolerant plants are well-adapted to such permanent abiotic stress. Surprisingly, they are able to survive and reproduce under environmental conditions of more than 200 mM NaCl concentration and are also able to live permanently in extreme conditions. The aim of this study was to identify characteristic and indicator plant species for permanently alkaline soda and non-alkaline saline wetland habitats, and to analyze the relationship between the chemical composition of water and the plant’s saturation index. Based on the obtained results we can conclude that plant species identified to live in high saline conditions may also grow under high alkaline conditions. Two groups were distinguished based on the salinity and alkalinity: “Saline group” includes all types of halophytes and halotolerant species, while “Soda group” includes obligate alkalophytes and alkalotolerant species. However, the diversity of plants in alkaline environment is bigger than in saline environment, and most of the halotolerant species grow in both alkaline soda and saline habitats. It is possible to identify three indicator species requiring alkaline soda habitats to live in (Aster tripolium, Puccinellia limosa, Suaeda pannonica) and three indicator species that dominantly grow in non-alkaline saline habitats (Juncus maritimus, Salicornia prostrata, Suaeda salsa). Different independent variables have been identified and analyzed to understand their influence on the species saturation index by calculating percentage of characteristic plant species of the entire region. It was identified that the plant saturation index has a significant positive relationship with the Mean Temperature of the Warmest Quarter, Topsoil Sand, and Clay fraction variables, and there is a significant negative relationship with the Solid Earth surface heat flow, as the heating can increase the extremity of the environment. Identification of indicator species of salinity and alkalization makes it possible to determine the type of soil, water bodies and habitats as a rapid biomonitoring method as a visual observation, without the use of expensive equipment and chemical laboratory tests.

Effect of DNA's Molecular Weight on their Solution Viscosity, Critical Concentrations, and Liquid Crystals Formation

AbstractThe study of the dynamics of nucleic acids in solution, their flow properties, and viscoelasticity is of great importance for understanding their biological functions. Many important properties of nucleic acids, such as DNA, depend on polymer concentration, CDNA, molecular weight (MW), rigidity, and external salt content, the latter parameter affecting electrostatic interactions. Furthermore, DNA molecular chains can organize, in vitro, into liquid crystalline phases at high CDNA. In this work, three DNA samples with different MW were studied in solution in a broad concentration range (CDNA from 0.025 to 200 mg mL−1, depending on DNA MW). Firstly, the intrinsic viscosities and MW for all DNA samples were determined through capillary and rheological measurements. Then, the overlap concentrations, C*, were estimated from the relation C* ≈ [η]−1. DNA chain characteristics were then analyzed in terms of the influence of DNA MW on the solution viscosities and on the overlap parameter, CDNA[η]. Flow birefringence appearance was identified by screening a wide CDNA range through visual observations with crossed polarizers. Finally, crossed‐light polarized microscopy was used to identify the appearance of liquid crystals at rest, confirming that higher CDNA are needed to obtain liquid crystals for low MW DNA samples.

Citrus leaf disease detection through deep learning approach

The majority of people in the world directly or indirectly depend on agriculture. Plant diseases are a significant threat to agricultural production and food security. Due to its high nutritional value, citrus fruit is one of the most abundant fruits in the world. However, different diseases are responsible for degraded citrus production as well as financial losses to the farmers. Traditionally, visual observation by experts has been attended to diagnose plant diseases. Usually, plant leaf disease recognition methods mainly rely on expert experiences to manually extract the colour, composition, and other features of diseased leaf images. Black spot, greening, canker, and melanoses are four common citrus leaf diseases. Rapid and accurate diagnosis of these diseases is a demand of time. Deep learning is a promising solution to these problems. There are different types of deep learning architecture like ImageNet, GoogleNet, VGG16, ResNet50, and InceptionV3, which show promising results in different object detection. Though most of these benchmark models give almost similar accuracy. However, this paper uses two deep learning models to find the better ones for the detection of citrus leaf disease detection. Hence, InceptionV3 outperforms VGG16 in terms of accuracy.

Removal of graffiti paint from construction materials coated with TiO2-based photocatalysts.

Graffiti on construction materials has significant social and economic impacts, especially on artistic and historical artefacts. Anti-graffiti protective coatings are used to generate low surface energies that limit graffiti adhesion to the surface, thereby reducing surface damage and facilitating removal. The anti-graffiti properties of three commercial TiO2-based coatings were tested under outdoor exposure conditions using four colours of graffiti paint (red, blue, black, and white). Chemical removers were used to clean the stained surfaces to understand the impact of the photocatalytic coatings during the conventional cleaning procedure. The effectiveness of cleaning was assessed by visual observations, colour measurements, and the percentage of residual stain. The anti-graffiti efficacy was strongly dependent on the colour of the graffiti and characteristics of the TiO2 coating. The cleaning performance of TiO2-treated samples was likely related to the photocatalytic redox reactions that decompose the graffiti. Additionally, their hydrophilicity may also prevent the adhesion and/or penetration of graffiti paint on the surface and/or pore matrix.

Development of a Tremor Detection Algorithm for Use in an Academic Movement Disorders Center.

Tremor, defined as an "involuntary, rhythmic, oscillatory movement of a body part", is a key feature of many neurological conditions including Parkinson's disease and essential tremor. Clinical assessment continues to be performed by visual observation with quantification on clinical scales. Methodologies for objectively quantifying tremor are promising but remain non-standardized across centers. Our center performs full-body behavioral testing with 3D motion capture for clinical and research purposes in patients with Parkinson's disease, essential tremor, and other conditions. The objective of this study was to assess the ability of several candidate processing pipelines to identify the presence or absence of tremor in kinematic data from patients with confirmed movement disorders and compare them to expert ratings from movement disorders specialists. We curated a database of 2272 separate kinematic data recordings from our center, each of which was contemporaneously annotated as tremor present or absent by a movement physician. We compared the ability of six separate processing pipelines to recreate clinician ratings based on F1 score, in addition to accuracy, precision, and recall. The performance across algorithms was generally comparable. The average F1 score was 0.84±0.02 (mean ± SD; range 0.81-0.87). The second highest performing algorithm (cross-validated F1=0.87) was a hybrid that used engineered features adapted from an algorithm in longstanding clinical use with a modern Support Vector Machine classifier. Taken together, our results suggest the potential to update legacy clinical decision support systems to incorporate modern machine learning classifiers to create better-performing tools.

Integrated gold nanorods-based dual-signal platform for accurate total antioxidant capacity assessment in food samples

Accurate assessment of Total Antioxidant Capacity (TAC) in food is crucial for evaluating nutritional quality and potential health benefits. This study aims to enhance the sensitivity and reliability of TAC detection through a dual-signal method, combining colorimetric and photothermal signals. Gold nanorods (AuNRs) were utilized to establish a dual-signal method duo to the colorimetric and photothermal properties. Fenton reaction can etch the AuNRs from the tips, as a result, a blue shift in the longitudinal LSPR absorption peak was obtained, leading to significant changes in color and photothermal effects, facilitating discrimination through both visual observation and thermometer measurements. In the presence of antioxidants, the Fenton reaction was suppressed or inhibited, protecting the AuNRs from etching. The colorimetric and photothermal signals were therefore positively correlated with TAC levels, enabling dual-signal detection of TAC. The linear range of AA was 4–100 μM in both colorimetry and photothermal modes, with detection limits of 1.60 μM and 1.38 μM, respectively. This dual-signal approach achieves low detection limits, enhancing precision and sensitivity. The method thus has the potential to act as a promising candidate for TAC detection in food samples, contributing to improved food quality and safety assessment.

Penilaian Spatial dan Temporal Gaya Berjalan dalam Satah Hadapan menggunakan MediaPipe Pose

Gait analysis is a vital field in biomechanics, focusing on studying human walking patterns and movement. Traditional methods relied on visual observation, lacking accuracy and objectivity. Instrumented gait analysis improved measurements using specialized equipment like motion capture systems and wearable sensors, but had limitations due to cost, and markers or sensors interfering with natural movement. The field underwent a significant transformation with the advent of markerless pose estimation-based gait analysis, which harnessed computer vision methodologies to monitor human movement without the requirement for specialized equipment, delivering a cost-effective and easily accessible analysis in real-world environments. However, it still faces challenges, including limitations in the sagittal plane and reliance on computationally demanding models like OpenPose. This study introduced a novel frontal plane gait analysis approach using the lightweight MediaPipe Pose model and a single camera setup. The objective is to evaluate the feasibility, and accuracy of MediaPipe Pose by comparing it to the established 3D Vicon motion capture system. The proposed approach tracked body keypoints during gait, detected gait events based on ankle depth changes and vertical difference between left and right ankles, and calculated gait parameters. The findings demonstrated that MediaPipe Pose-based gait analysis showed promise for accurately analyzing gait parameters in the frontal plane with low mean absolute error (0.00–0.30). Combining this method with satellite technology enhances e-health by facilitating teleconsultations, remote diagnostics, and extending healthcare services to disaster-stricken or remote areas.

Black-hole and DoS attack detection analysis and isolation mechanism for black-hole in MANET

A mobile ad-hoc network is a network without infrastructure where communication between nodes occurs in an open medium, such as the air. This makes it susceptible to eavesdropping and various attacks, which can be categorized as data traffic attacks (such as black-hole, grey-hole, DoS, etc.) and control traffic attacks (such as worm-hole, hello flood, man-in-the-middle, etc.). This NS2 experimental simulation compares the disruption effects of black-hole and DoS attacks in an identical network using the AODV routing protocol. The simulation aims to observe and analyze parameters such as packet delivery fraction, end-to-end delay, normalized overhead, and residual energy in HKI-Reader. Additionally, it demonstrates how malicious nodes are detected, removed, or isolated from the network to restore it to its original state. HKI-Reader is a software application developed using PHP and JavaScript. It is designed to read simulation results files for attacks and display them in a graphical user interface (GUI) for improved analysis and visual observation.

Application of Digital Cameras Toward Children Development Domain

This research was carried out to create awaereness on the benefits of camera applications to young children development like social, language, and cognitive development. The initiative to introduce camera as teaching tool in preschool and home was one of fantastic ideas in effort of making children’s learning more interactived and improved. Digital cameras provide children with a medium to express their creativity, capture and share moments, and engage in storytelling. This studies proved that the digital camera seccussfuly encouraged children to observe and appreciate the world around them, enhancing their visual literacy and observation skills. Furthermore, digital cameras foster social interactions and collaboration as children work together on photography projects, share their creations, and connect with others who share their interests.

Optimization of CRISPR/Cas12a detection assay and its application in the detection of Echinococcus granulosus

Cystic echinococcosis, resulting from infection with Echinococcus granulosus, poses a significant challenge as a neglected tropical disease owing to the lack of any known effective treatment. Primarily affecting under-resourced, remote, and conflict-ridden regions, the disease is compounded by the limitations of current detection techniques, such as microscopy, physical imaging, ELISA, and qPCR, which are unsuitable for application in these areas. The emergence of CRISPR/Cas12a as a promising tool for nucleic acid detection, characterized by its unparalleled specificity, heightened sensitivity, and rapid detection time, offers a potential solution. In this study, we present a one-pot CRISPR/Cas12a detection method for E. granulosus (genotype G1, sheep strain) integrating recombinase polymerase amplification (RPA) with suboptimal protospacer adjacent motif (PAM) and structured CRISPR RNA (crRNA) to enhance reaction efficiency. The evaluation of the assay's performance using hydatid cyst spiked dog feces and the examination of 62 dog fecal samples collected from various regions of Western China demonstrate its efficacy. The assay permits visual observation of test results about 15 minutes under blue light and displays superior portability and reaction speed relative to qPCR, achieving a sensitivity level of 10 copies of standard plasmids of the target gene. Analytic specificity was verified against four tapeworm species (E. multilocularis, H. taeniaeformis, M. benedeni, and D. caninum) and two other helminths (T. canis and F. hepatica), with negative results also noted for Mesocestoides sp. This study presents a rapid, sensitive, and time-efficient DNA detection method for E. granulosus of hydatid cyst spiked and clinical dog feces, potential serving as an alternative tool for field detection. This novel assay is primarily used to diagnose the definitive host of E. granulosus. Further validation using a larger set of clinical fecal samples is warranted, along with additional exploration of more effective approaches for nucleic acid release.

In Situ Visual Observation of Surface Energy-Controlled Heterogeneous Nucleation of Metal Nanocrystals.

Electrochemical growth of metal nanocrystals is pivotal for material synthesis, processing, and resource recovery. Understanding the heterogeneous interface between electrolyte and electrode is crucial for nanocrystal nucleation, but the influence of this interaction is still poorly understood. This study employs advanced in situ measurements to investigate the heterogeneous nucleation of metals on solid surfaces. By observing the copper nanocrystal electrodeposition, an interphase interaction-induced nucleation mechanism highly dependent on substrate surface energy is uncovered. It shows that a high-energy (HE) electrode tended to form a polycrystalline structure, while a low-energy (LE) electrode induced a monocrystalline structure. Raman and electrochemical characterizations confirmed that HE interface enhances the interphase interaction, reducing the nucleation barrier for the sturdy nanostructures. This leads to a 30.92-52.21% reduction in the crystal layer thickness and a 19.18-31.78% increase in the charge transfer capability, promoting the formation of a uniform and compact film. The structural compactness of the early nucleated crystals enhances the deposit stability for long-duration electrodeposition. This research not only inspires comprehension of physicochemical processes correlated with heterogeneous nucleation, but also paves a new avenue for high-quality synthesis and efficient recovery of metallic nanomaterials.

Accuracy of geometric morphometrics for age estimation using frontal face photographs of children and adolescents: A promising method for forensic practice

Age estimation is crucial in legal and humanitarian contexts. Forensic professionals may use various procedures to estimate age, including dental analysis, bone density tests, evaluation of physical characteristics including facial bone structure and development, and image-based methods. Although images are often the only material available, visual observation of photographic material is an imprecise method in age estimation, which can compromise judicial decision-making. Analyzing 4000 photographs from the Brazilian Federal Police database, representing four age groups (6, 10, 14, and 18 years), the study employed automated analysis by marking 28 photogrammetric points. Data were used to establish facial patterns by age and sex using the facial geometric morphometrics method. Performance was assessed through a Multinomial Logistic Regression model, evaluating accuracy, sensitivity, and specificity across the categorical age groups. Analyses were conducted using R software, with a 5 % significance level. The study found that facial geometric morphometrics achieved an overall accuracy of 69.3 % in age discrimination, with higher accuracy in males (74.7 %) compared to females (65.8 %) (p < 0.001). The method excelled at predicting the age of 6-year-olds with 87.3 % sensitivity and 95.6 % specificity but had lower performance at 14 years. It showed greater accuracy in distinguishing age groups with larger age gaps, achieving up to 99.5 % accuracy between certain groups, and was particularly effective in differentiating ages of 6 and 10 years in females and 10, 14, and 18 years in males. The facial geometric morphometrics emerges as a promising approach for age estimation among children and adolescents in forensic settings.

Infrared emissivity of icy surfaces

Context. Most analyses of the infrared emission of Saturn’s rings and icy satellites have considered pure water ice as the constituent of regolith and particle surfaces. Visual and near-infrared observations have shown, however, that darkening and reddening contaminants are present at a fraction level of a few percent. In the spectral domain 10–2000 cm−1, water ice becomes transparent in a few windows, which in particular causes the roll-off of emissivity of icy surfaces that is observed below 50 cm−1. Their emissivity there may be affected by these contaminants. Aims. We present a quantitative global sensitivity analysis of a hybrid Mie-Hapke model to evaluate the influence of regolith properties and contaminant fraction on the infrared emissivity of icy rings or moons over this spectral range. Methods. A hybrid Mie–Hapke model of the hemispherical emissivity ε*h(Wn) was made, including various diffraction correction and mixing types with tholins or amorphous carbon grains, or grain size distributions and some anisotropy in emission. A Sobol global sensitivity analysis provided quantitative levels of importance for these factors versus wave number wn. Results. Given the a priori uncertainties, the most important factor acting on ε*h(Wn) remains the size distribution of regolith grains and the average anisotropy factor ξ. For wn&gt; 50 cm−1, ξ, the power-law index p and the minimum amin of the size distribution are most influential. In windows of water-ice transparency (10–50, 300–600, and 900–1300 cm−1), the emissivity is also sensitive, but to a lesser extent, to the maximum grain size amax and the fraction f of contaminants, if mixed at the molecular level. Conclusions. This model provides a self-consistent tool for interpreting multi-modal observations of the thermal emission from icy surfaces. It also offers interesting insights into recent mid-infrared observations of Saturn’s rings and Jupiter’s moon Ganymede by the JWST-MIRI instrument.

Adaptive Sliding Mode Trajectory Tracking Control of Unmanned Surface Vessels Based on Time-Domain Wave Inversion

In this work, we develop a trajectory tracking control method for unmanned surface vessels (USVs) based on real-time compensation for actual wave disturbances. Firstly, wave information from the actual sea surface is extracted through stereoscopic visual observations, and data preprocessing is performed using a task-driven point cloud downsampling network. We reconstruct the phase-resolved wave field in real time. Subsequently, the wave disturbances are modeled mechanically, and real-time wave disturbances are used as feedforward inputs. Furthermore, an adaptive backstepping sliding mode control law based on command filters is designed to avoid differential explosion and mitigate sliding mode chattering. An adaptive law is also designed to estimate and compensate for other external disturbances and inversion error bounds that cannot be computed in real time. Finally, the feasibility of the proposed control strategy is validated through stability analysis and numerical simulation experiments.

Online Surveillance of IoT Agents in Smart Cities Using Deep Reinforcement Learning

In the context of today's smart cities, the effective operation of online surveillance of IoT agents is crucial for maintaining public safety and security. To achieve this, collaboration and cooperation among these autonomous IoT agents are indispensable. While the existing research has focused on collaboration amongst the neighboring agents or implicit cooperation, real-world scenarios often necessitate broader forms of collaboration. In response to this need, we introduce a novel framework that leverages visual signals and observations to facilitate collaboration among online surveillance. Our proposed framework incorporates the Multi-Agent POsthumous Credit Assignment (MA-POCA) algorithm as a training mechanism. The empirical results demonstrate that our framework consistently outperforms the base model in various performance metrics. Specifically, it exhibits superior performance in group cumulative reward, cumulative reward, and episode length. Furthermore, our proposed model excels in policy loss performance measures when compared to base model.