Olfactory implants to address anosmia have gained interest in recent years. Existing transnasal and transcranial approaches to the olfactory bulb (OB) have potential complications. We aim to determine whether transcranial supraorbital keyhole craniotomy (SOKC) provides adequate and safe access to the OB. Secondary outcomes include highlighting specific anatomical obstructions and impacts of patient characteristics. This is a retrospective cohort study. This study was conducted at the Tertiary Academic Medical Center. Fifty fine-cut computed tomographies of the sinuses in consecutive adult patients (50% male) were analyzed. Image processing was performed using syngo.via to assess whether there was a clear path from points on the exterior skull to the anterior or posterior OB. Using five points based on the SOKC technique on the exterior skull, lines were drawn from each point to the ipsilateral anterior and posterior OB resulting in 1,000 pathways. Pathways were reconstructed and analyzed for violations of the orbit, orbital bone, or sinuses. A total of 96% of the subjects had at least one unobstructed pathway to the OB. The route most commonly unobstructed (90%) was 2P (2 cm above the supraorbital notch to posterior bulb). The posterior OB was less obstructed than the anterior (56, 112; p = 0.00002). The most common obstruction was an ipsilateral sinus. The SOKC may be an effective and safe approach for an olfactory implant in many patients. The posterior OB had a clearer approach with less obstruction. By avoiding injury to the nasal mucosa, orbit, sinuses, or traversing the skull base, this approach may prove advantageous over transnasal approaches for olfactory implantation in select patients.

High resolution 3D model-based topographic assessment of scour around a hollow artificial reef by integration monocular vision and image processing techniques

Measurement of the proton water-equivalent path length (WEPL) and relative stopping power (RSP) of human tissues with proton radiography (pRG) can help improving the X-ray-CT-based RSP images of patients in proton-therapy treatment plans. This paper presents an extension to cylindrical symmetric two-material phantoms of a pRG method based on one single 2D-pixelated flat detector, previously applied by us to homogeneous materials. The experiments are simulated with Monte-Carlo simulations and performed with a scintillator screen coupled with a CCD camera. The spatial 2D distributions of the integral energy deposit and proton fluence in the detector after each phantom are determined. In the experiments, only the energy deposit is measured (by the intensity of the scintillation light from the screen), whereas the simulated fluence distribution is used as a proxy for the experimental counterpart. The ratio of the two distributions is calculated, which (in first instance) only depends on the stopping power of protons in the detector. This ratio is converted to an image of the residual range in water of protons after traversing the phantom with a calibration, which also involves correcting for the energy deposit from particles not determining the range values, e.g. protons having undergone hard interactions and secondary particles. The range image is then converted to a WEPL/RSP image, and the measured WEPLs/RSPs are compared with reference values to assess their accuracy. Simulations showed that 1 % accurate RSPs can be achieved for the homogeneous parts of the phantoms, whereas this is hampered near the border of different materials by range mixing due to proton multiple Coulomb scattering (MCS). The results from the experiments are overall consistent with the simulations, but 1 % accurate RSPs were not measured for some parts of the phantoms because of image artefacts related to the use of the simulated fluence distribution instead of the experimental one. A novel image-processing methodology for pRG with a scintillator screen was investigated. The WEPL/RSP of a target is determined from the ratio of the energy deposit and fluence of protons in the screen through a designated calibration. To test the WEPL/RSP accuracy of the method, the pRGs of two phantoms with simple geometry and made of two different homogeneous materials were simulated and measured. The simulations showed that the RSP can be determined with 1 % accuracy over all homogeneous parts, whereas this is hampered at the border between different materials by MCS-induced range mixing, which needs more complicated analysis to be disentangled. In the experiments, 1 % accurate RSPs could be measured only over a limited area of the image because of artefacts caused by using the simulated fluence distribution in place of the experimental one, which must then be measured in future experiments. • Proton radiography of two-material phantoms with a scintillator screen. • Novel image-processing method used to retrieve the RSP image. • In simulations, 1 % accurate RSPs except at the border between different materials. • Measuring the proton fluence experimentally is fundamental to avoid artefacts.

Automated design framework for concrete planar components: Integrating topology optimization, strut-and-tie model, and image processing techniques

Image Processing-Based Automatic Tooth Segmentation and Age Estimation in Sheep Using Deep Learning.

Image processing techniques for viewpoint correction and resolution enhancement in light field 3D displays

Dynamic dilemma zone at signalized intersection: attention allocation patterns using cure survival analysis for male riders.

Brain tumor detection using HyGSNet and feature extraction with DWT-based GDP.

Artificial intelligence in animal anatomy: Exploring the technologies, applications, benefits, and challenges.

Decoding Sumerian craft technologies: morphological image processing and mesoscopic feature analysis of archaeological bitumen-based composites

Non-destructive monitoring of postharvest quality changes in chili (Capsicum frutescens) during storage using image processing coupled with machine learning

Real projection algorithms for generalized low-rank approximation of large-scale quaternion matrix in color image processing

Pisang Island on Pesisir Barat has great potential to be developed as a marine ecotourism destination, particularly due to its coral reef ecosystem, seagrass beds, and beautiful, picturesque beaches. However, when developing this tourism, it is important to consider the balance between ecosystem protection and the economic needs of coastal communities, so that the ecosystem remains sustainable. This study aims to determine sustainable marine ecotourism zones using Marxan. We integrated ecological variables— bathymetry, brightness, current velocity, coral reefs and seagrass cover beds—with social layers: jetties, boat sailing lanes, traditional fishing areas, and surfing spots. The integrated ecological data includes bathymetric maps from BATNAS, water brightness indices from Sentinel-2A image processing, surface current velocities from Copernicus Marine Service, and maps of coral reefs and seagrass bed distribution from Allen Coral Atlas. Socio-economic data was obtained by identifying jetties locations, boat sailing lanes, traditional fishing areas, and established tourist sites around Pisang Island. Three habitat protection targets (30%, 40%, 50%) were tested through Marxan Zoning on hexagonal planning units. The model results are displayed using the summedsolution frequency approach to see how often a unit is selected. After that, the zones are divided into four categories: core zone, buffer zone, marine ecotourism zone, and other zone. The results of the study indicate that Scenario A, with a protection target of 30%, is the most optimal choice, with an allocation of 120,670 ha (5.30%) for the ecotourism zone and 26,199 ha (1.15%) for the core zone. Scenario A is the most balanced approach in supporting marine ecosystem conservation and coastal community empowerment through tourism activities. This approach is considered the most balanced as it preserves marine ecosystems while providing space for tourism that supports local community economic benefits. These findings highlight the important role of Marxan as a data-driven spatial planning tool to support sustainable marine ecotourism management policies in Pisang Island.

A portable multiplex vertical flow immunoassay platform for Tier 1 biothreat detection in biofluids and environmental matrices.

In this study, an image processing-based deep learning approach is developed to classify the filler ratios in glass fiber reinforced polymer composite with hybrid MgO-CuO nanoparticles in a non-contact and non-destructive manner, based solely on surface color and texture information. The originality of the study lies in its reliance on the direct learning of hybrid nanoparticle dopant ratios from optical surface properties, unlike image-based methods in the literature that mainly focus on damage detection, phase separation, or mechanical property estimation. In this context, eight different composite classes with different MgO and CuO weight ratios were produced and the samples were imaged at high resolution under homogeneous LED illumination at a fixed camera-sample distance. The obtained images were evaluated in a multi-class classification problem using the transfer learning-based EfficientNet-B0 architecture without any data enhancement. Model performance was analyzed with accuracy, sensitivity, recall, and F1-score metrics. Test results show that the proposed model achieved 97% overall accuracy and a macro-mean F1-score of 0.97; The study demonstrated that single-dopant and high-contrast hybrid systems were classified with high accuracy rate. Limited class overlap was observed in some hybrid classes with low CuO content, which is attributed to the optical similarity of the dopants. The findings reveal that deep learning approaches based on surface images offer a powerful, low-cost, and industrially viable alternative for dope ratio verification and rapid quality control applications in composite manufacturing processes.

ABSTRACT Marble, widely used in heritage structures, is highly susceptible to durability issues, with staining being a major concern. Choosing the optimal destaining approach is vital in preserving the aesthetic and structural integrity of marble surfaces, as repeated interventions can have long-term effects. However, traditional evaluation methods rely either on subjective visual inspection, leading to inconsistencies and on the sophisticated laboratory analyses requiring extensive resources. This study develops a MATLAB-based image analysis protocol to quantitatively assess the efficiency of destaining techniques for marble surfaces. By integrating advanced image processing algorithms, such as stain detection, preprocessing, and quantification metrics, the protocol provides a reproducible framework for evaluating cleaning effectiveness. The results are validated using the microanalytical tools, such as Scanning Electron Microscopy and Energy Dispersive Spectroscopy, to assess destaining efficiency by mineralogical mapping and quantification. The destaining process employed was poulticing, utilizing an optimum combination of cellulose powder and clay in ratio 9:1. This study paves a way to quantify stain removal efficiency through image analysis by MATLAB coding, and also focuses on improving the poulticing methodology using cellulose-clay combination to effectively remove various types of stains. Together, these efforts support better practice for the conservation of marble structures.

Structural cracks in concrete infrastructures such as buildings, bridges, and pavements pose significant safety and durability risks if not detected at an early stage. Conventional crack inspection methods rely heavily on manual visual assessment, which is time-consuming, subjective, and unsafe for large-scale or hard-to-reach structures. To address these limitations, this project presents a Drone-Based Concrete Crack Detection System using Machine Learning, integrated with an interactive Streamlit web application. The system utilizes drone-captured images as input and employs a Convolutional Neural Network (CNN) to automatically detect the presence of cracks. In addition, image processing techniques such as grayscale conversion, thresholding, and pixel analysis are applied to estimate crack severity quantitatively. The developed web-based interface allows users to upload multiple images, visualize detected cracks through overlay highlighting, and receive automated decision outputs. The system also generates downloadable CSV and PDF reports and provides audio-based result announcements to enhance usability. The proposed solution offers an efficient, scalable, and non-invasive approach for automated structural health monitoring and early crack detection.

Research on characteristic recognition and quantification of internal powder residue in LPBF porous structure based on image processing.

Bronchopulmonary dysplasia (BPD) is a common complication among preterm neonates, with portable X-ray imaging serving as the standard diagnostic modality in neonatal intensive care units (NICUs). However, lung magnetic resonance imaging (MRI) offers a non-invasive alternative that avoids sedation and radiation while providing detailed insights into the underlying mechanisms of BPD. Leveraging high-resolution 3D MRI data, advanced image processing and semantic segmentation algorithms can be developed to assist clinicians in identifying the etiology of BPD. In this dataset, we present MRI scans paired with corresponding semantic segmentations of the lungs and trachea for 40 neonates, the majority of whom are diagnosed with BPD. The imaging data consist of free-breathing 3D stack-of-stars radial gradient echo acquisitions, known as the StarVIBE series. Additionally, we provide comprehensive clinical data and baseline segmentation models, validated against clinical assessments, to support further research and development in neonatal lung imaging.

The objective of this study was to analyze leaf area data for four cotton cultivars estimated using the Petiole and LeafArea software, with leaf area calculated as the product of leaf length, leaf width, and a correction factor, and the LI-COR device method used as the standard for validation. The validation of the data estimated by the Petiole software and LeafArea, and the method by which leaf length and width correlate, was performed using statistical indices to evaluate precision and accuracy. The petiole showed the best performance in estimating leaf area for the cultivars TMG44B2RF, IMA5801B2RF, and FM985GLTP. The methodology using leaf dimension measurements and a correction factor is more effective for estimating the leaf area of cultivar BASF – FM 944GL. Among all the evaluated methods, the LeafArea method had the lowest effect on the leaf area of the cotton cultivars. The methodologies evaluated provide precise and accurate estimates of leaf area for the different cotton cultivars, with results that vary across cultivars. Keywords: leaf morphometry; digital photography; Gossypium hirsutum L.; image processing.