Sorting Process Research Articles

Precision isolation and cultivation of single cells by vortex and flat-top laser ejection.

Single-cell isolation stands as a critical step in single-cell studies, and single-cell ejection technology based on laser induced forward transfer technology (LIFT) is considered one of the most promising methods in this regard for its ability of visible isolating single cell from complex samples. In this study, we improve the LIFT technology and introduce optical vortex laser-induced forward transfer (OV-LIFT) and flat-top laser-induced forward transfer (FT-LIFT) by utilizing spatial light modulator (SLM), aiming to enhance the precision of single-cell sorting and the cell's viability after ejection. Experimental results demonstrate that applying vortex and flat-top beams during the sorting and collection process enables precise retrieval of single cells within diameter ranges of 50 μm and 100 μm, respectively. The recovery rates of Saccharomyces cerevisiae and Escherichia coli DH5α single cell ejected by vortex beam are 89 and 78%, by flat-top beam are 85 and 57%. When employing Gaussian beam sorting, the receiving range extends to 400 μm, with cultivation success rates of S. cerevisiae and E. coli DH5α single cell are 48 and 19%, respectively. This marks the first application of different mode beams in the ejection and cultivation of single cells, providing a novel and effective approach for the precise isolation and improving the viability of single cells.

Parameters optimization for decontamination and fine physical regeneration pathways of polypropylene plastics from waste lunchboxes

Due to the development of the food delivery industry, a large amount of waste lunchboxes made of homo polypropylene (PP) plastic have been generated. This study developed a new technological strategy to effectively regenerate PP from waste lunchboxes. Through response surface curve analysis, it was found that under the optimal process conditions of hot alkali washing at 80 ℃, 30 min, and pH 13, the optimal contact angle was 65.55°, indicating a good oil stain removal effect. By identifying and analyzing the characteristics of impurities in waste lunchboxes, a physical sorting and granulation regeneration process was constructed. And through large-scale statistical analysis and data collection, it was further verified that recycled PP plastics maintained their physical stability and excellent processing performance. The quality stability of recycled PP plastics in terms of impurities content was also verified. By designing different formulations specifically, recycled PP was mixed with different virgin PP and antioxidants in appropriate proportions, and extruded into particles under 150–300 mesh filtration conditions to obtain modified recycled PP. Modified recycled PP was applied in textiles, clothing, and injection molded products. In conclusion, we achieve the up-cylcing of waste PP lunchboxes instead of down-cylcing.

Cutout as augmentation in contrastive learning for detecting burn marks in plastic granules

Abstract. Plastic granules are a common delivery form for creating products in industries such as the plastic manufacturing, construction and automotive ones. In the corresponding sorting process of plastic granules, diverse defect types could appear. Burn marks, which potentially lead to weakened structural integrity of the plastic, are one of the most common types. Thus, plastic granules with burn marks should be filtered out during the sorting process. Artificial intelligence (AI)-based anomaly detection approaches are widely used in the field of visual-based sorting due to the higher accuracy and lower requirement of expert knowledge compared with classic rule-based algorithms (Chandola et al., 2009). In this contribution, a simple data augmentation strategy, cutout, is implemented as a way of simulating defects when combined with a contrastive learning-based methodology and is proven to improve the accuracy of the anomaly detection of burn marks. Different variants of cutout are also evaluated. Specifically, synthetic image data are used due to the lack of real data.

The nature, causes, and consequences of inter-regional inequality

Abstract Social scientists and policymakers alike have become increasingly concerned with understanding the nature, causes, and consequences of inter-regional inequality in economic living conditions. Contemporary spatial inequality is multi-faceted—it varies depending on how we define inequality, the scale at which it is measured, and which groups in the labor force are considered. Increasing economic inequality has important implications for broader social and political issues. Notably, it is difficult to account for the rise of far-right populism in industrialized countries without considering the context of growing inter-regional inequality. Important explanations for the rise in inter-regional inequality include changing patterns of worker and firm sorting processes across space, major transitions like the reorientation of the economy from manufacturing to digital technologies, and increasing global economic integration, as well as policy. Different causal explanations in turn imply a different role for place-based policy. This article introduces the context of the special issue on the nature, causes, and consequences of inter-regional inequality, focusing specifically on inequality in North America and Western Europe, and aims to identify challenges for, and spark further research on, inter-regional inequality.

Physicochemical Characterization of Simplicia Sambiloto (Andrographis paniculata) Powder with Types and Concentrations of Reagents in the Formation of Zn-Chlorophyll Complexes

During the process of making sambiloto simplicial which involves heat, it can cause degradation of chlorophyll. Increasing the stability of chlorophyll can be done by forming Zn-chlorophyll complexes with metals which are be able to form complexes where more stable than Mg such as by using Zn. The aim of this research is to determine how the physical and chemical characteristics of the bitter simplicia powder produced are influenced by the type of reagent and the concentration of the reagent when heated in an oven. Simplicia bitter powder is made through a sorting, grinding and sieving process. Next, the zinc chloride and zinc acetate reagents were mixed with concentrations of 0, 200, 300, 400, and 500 ppm, and then dried. The results showed a real effect (p value <0.05) of the interaction between the types of ZnCl2 and Zn acetate reagents and the reagent concentration on color intensity, color stability, water content, ash content and chlorophyll content. The higher the concentration of ZnCl2 and Zn acetate, the higher the water content, ash content, chlorophyll content, L*, and -a*, while b* and ∆E decrease. At a concentration of 500 ppm ZnCl2 showed an increase in water, ash, chlorophyll, ∆E, L*, and a* content while b* decreased. Meanwhile, Zn acetate 500 ppm moisture content, ash content, chlorophyll content, L*, and ∆E decreased, while a* and b* increased. Sambiloto simplicia powder which has the highest chlorophyll content is Zn acetate 400 ppm, with characteristics of water content 6.81% ww, ash content 15.74% ww, total chlorophyll 397.97 mg/100 g ww, green color intensity (-a) -2.93, and ∆E 5.48.

Evaluating the Performance of Voting Classifier in Multiclass Classification of Dry Bean Varieties

This study explores the application of a voting classifier, integrating Decision Tree, Logistic Regression, and Gaussian Naive Bayes models, for the multiclass classification of dry bean varieties. Utilizing a dataset comprising 13,611 images of dry bean grains, captured through a high-resolution computer vision system, we extracted 16 features to train and test the classifier. Through a rigorous 5-fold cross-validation process, we assessed the model's performance, focusing on accuracy, precision, recall, and F1-score metrics. The results demonstrated significant variability in the classifier's performance across different data subsets, with accuracy rates fluctuating between 31.23% and 96.73%. This variability highlights the classifier's potential under specific conditions while also indicating areas for improvement. The research contributes to the agricultural informatics field by showcasing the effectiveness and challenges of using ensemble learning methods for crop variety classification, a crucial task for enhancing agricultural productivity and food security. Recommendations for future research include exploring additional features to improve model generalization, extending the dataset for broader applicability, and comparing the voting classifier's performance with other ensemble methods or advanced machine learning models. This study underscores the importance of machine learning in advancing agricultural classification tasks, paving the way for more efficient and accurate crop sorting and grading processes.

A Qualitative Focus on School Leaders' Perceptions of Roles, Responsibilities, and Expectations in Enhancing Equitable Access to Educational Technology In the Era of Gen-AI

The purpose of this study is to provide a phenomenological perspective on school leaders' perceptions of their roles, responsibilities and expectations in increasing equal access to educational technology in the Age of Artificial Intelligence. This research was conducted through in-depth interviews with a phenomenological approach. The participants consisted of school leaders (school principals [n = 24] and vice principals [n = 18]) working at various educational levels. The qualitative data obtained were analysed by content analysis. Following the data collection process, themes were formed with main words or sentences as a result of descriptive processes. The data obtained through interviews were analysed in a three-stage process of sorting, coding and categorisation. According to the findings, school leaders perceive the responsibility of integrating productive AI technologies into their schools as strengthening the technological infrastructure and supporting R&D studies to ensure equal opportunities in education. In providing equal access to Productive AI educational technologies, the leaders stated that especially economic and socio-economic conditions should be improved and the importance of the correct and efficient use of these technologies. Recommendations for future research are presented.

Low-cost parallel delta robot for a pick-and-place application with the support of the vision system

The main topic of this paper is how to develop pick and place delta robot with high efficiency mechanical parts and prosaically assemble it, additionally built electrical wiring diagram all of this supported by vision computer system. The system has a low-cost part and solves many problems to build this robot. Electrical section has two main components master and slave the master is raspberry pi and slave is Arduino, master is used to vision detect object and Arduino is used to invers kinematics equation, move motors to target position, control in hydraulic system to pull the object with suction cup. The process of vison system is passing through some steps, detect the object by color range, contouring box then send the real position to an Arduino. The experimental process proved the effectiveness and accuracy of the delta robot vision system in the sorting process from conveyor system. The target of system that function is applied with a low cost.

Abstract 2530: Advancing ICK assay: Real-time, label-free imaging of lymphocyte subsets

Abstract Background: Immune Cell Killing (ICK) Assay is pivotal in immunology and clinical research, uncovering the intricate connections between the immune system and diseases. Traditional ICK assays utilizing antibody staining for endpoint data generation pose challenges in capturing dynamic interactions. Moreover, antibody labeling for data quantification may compromise sensitivity, especially with low-affinity antibodies or a limited number of cells. The anticipation of label-free lymphocyte subset classification and real-time effector/target cell interaction analysis in one field of view is expected to provide more valuable insights than traditional ICK assays. Previous studies achieved 70-80% accuracy in image-based lymphocyte subset classification using digital holographic microscopy, light scattering, etc. but the cells’ condition during the cell sorting or sample preparation process may affect the lymphocyte subset status, resulting in real morphological discrepancies. Our approach focuses on directly sorting human PBMCs to enable label-free, dynamic interaction analysis of natural immune system responses in one field of view. Method: In this study, PBMC samples from 6 individuals were separated using standard-density gradient centrifugation and labeled with antibodies. Data were acquired using the Holotomographic Microscope HT-X1 (Tomocube Inc.). We employed Densenet 121 as the deep learning model architecture, considering the trade-off between memory consumption and performance. The cross-entropy loss function was utilized, and the AdamW optimizer was chosen for model parameter optimization. Input data consisted of 3D refractive index (RI) data for individual cells, each of sized (20, 54, 54). Individual cell labeling was performed using staining information. Results: An overall accuracy of 93.75% in Human PBMC subtype cell classification was achieved. In stage 1, an accuracy of 97.56% was demonstrated in classifying [CD14, CD15, Others]. Among the cells classified as "Others" in stage 2, [CD3, CD19, CD16&CD56] displayed an accuracy of 93.5%. In stage 3, classifying [CD4, CD8] within the CD3 subtype achieved an accuracy of 90.2%. This classification is presumably based on pattern recognition of the chromosomal landscape in holotomography images. Conclusion: Our research has successfully demonstrated the feasibility of real-time label-free lymphocyte subset classification. This approach enabled the classification of effector cells as live cells in ICK assay and the analysis of effector-target cell dynamic interactions. Further training in the classification of more detailed subtype cells and various cell death types will provide additional insights. The advancement of this technology is expected to play a pivotal role in disease prevention, treatment, vaccine development, medical research, and academic studies. Citation Format: Sanggeun Oh, Jaephil Do, Hyun-Seok Min, Dongmin Ryu, Wei sun Park. Advancing ICK assay: Real-time, label-free imaging of lymphocyte subsets [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2530.

Data Fusion of RGB and Depth Data with Image Enhancement.

Since 3D sensors became popular, imaged depth data are easier to obtain in the consumer sector. In applications such as defect localization on industrial objects or mass/volume estimation, precise depth data is important and, thus, benefits from the usage of multiple information sources. However, a combination of RGB images and depth images can not only improve our understanding of objects, capacitating one to gain more information about objects but also enhance data quality. Combining different camera systems using data fusion can enable higher quality data since disadvantages can be compensated. Data fusion itself consists of data preparation and data registration. A challenge in data fusion is the different resolutions of sensors. Therefore, up- and downsampling algorithms are needed. This paper compares multiple up- and downsampling methods, such as different direct interpolation methods, joint bilateral upsampling (JBU), and Markov random fields (MRFs), in terms of their potential to create RGB-D images and improve the quality of depth information. In contrast to the literature in which imaging systems are adjusted to acquire the data of the same section simultaneously, the laboratory setup in this study was based on conveyor-based optical sorting processes, and therefore, the data were acquired at different time periods and different spatial locations. Data assignment and data cropping were necessary. In order to evaluate the results, root mean square error (RMSE), signal-to-noise ratio (SNR), correlation (CORR), universal quality index (UQI), and the contour offset are monitored. With JBU outperforming the other upsampling methods, achieving a meanRMSE = 25.22, mean SNR = 32.80, mean CORR = 0.99, and mean UQI = 0.97.

Semen sexing and its impact on fertility and genetic gain in cattle.

Semen sexing is among one of the most remarkable inventions of the past few decades in the field of reproductive biotechnology. The urge to produce offspring of a desired sex has remained since traditional times. Researchers have tried many methods for accurate semen sexing, but only the flow cytometry method has proved to be effective for commercial utilization. However, there were always concerns about the effects of sexed semen, especially on fertility and the rate of genetic gain. Some concerns were genuine because of factors such as low semen dosage in sexed semen straws and damage to sperm during the sorting process. Various researchers have conducted numerous studies to find out the effect of sexed semen on fertility and, in this article, we reflect on their findings. Initially, there were comparatively much lower conception rates (∼70% of conventional semen) but, with refinement in technology, this gap is bridging and the use of sexed semen will increase over time. Concerning genetic gain with use of sexed semen, a positive effect on rate of genetic progress with the use of sexed semen has been observed based on various simulation studies, although there has been a mild increase in inbreeding.

Analysis of Screw Press Machine Performance Using the Overall Equipment Effectiveness Method at PT Agro Synergy Nusantara

PT Agro Sinergi Nusantara PKS Aceh Seujahtera is a company engaged in the plantation and processing of oil palm fruit, where the production is in the form of crude palm oil (CPO), palm kernel and fiber. This factory has 2 horizontal type sterilizers with a capacity of 40 tons each. One machine that has a vital role in the production process is the screw press. Based on the results of observations and interviews with the company, it was found that the screw press machine was often damaged in the production process, as a result the production process was stopped and production results were not optimal, besides that the company would also experience losses. The purpose of this research is to measure the performance of the Screw Press Machine at PT Agro Sinergi Nusantara by calculating OEE through 3 main factors, namely Availability Ratio, Performance Effeciency Ratio, and Quality Ratio and determining corrective actions that can improve the performance of the Screw Press machine. The results of the performance analysis on the screw press machine obtained an Availability Ratio value of 84.52%, Performance Effeciency Ratio of 86.68%, Quality Ratio of 100%, and OEE of 73.37%. Based on a comparison with the Word Class Ideal Standard, only the Quality Ratio value meets the standard while the Availability Ratio, Performance Effeciency Ratio, and OEE have not met, meaning that the overall performance of the Screw Press Machine has decreased and has not met the Word Class Ideal Standard and it is necessary to improve the performance of the machine. Recommendations for improvements that can be made to improve the performance of the screw press machine again are to carry out routine maintenance on machine components so that damage can be prevented before severe damage occurs and cause longer downtime, conduct training for workers so that they can develop themselves and know more about the machine and the responsibilities that must be carried out, reinforce the implementation of routine and scheduled maintenance (preventive maintenance), and increase supervision during the FFB sorting process.

Finding a needle in a haystack: DNA Haemoproteus columbae enrichment using percoll density gradient and flow cytometry

Isolation of genomic DNA of blood parasites in birds, reptiles, amphibians, and fishes is a challenging task, given that their red blood cells are nucleated; for that reason, parasite genomic DNA is only a fraction of the total extracted DNA, and it is challenging to obtain concentrated high-quality genetic material. Percoll Density Gradient (PDG) and flow cytometry are tools for separating and analyzing cell populations or even a single cell, and both represent potent approaches for isolating avian haemosporidians parasites. Our experimental design included several steps seeking to concentrate the parasite´s DNA. We used blood samples from a Rock pigeon infected with Haemoproteus columbae. After inducing parasite exflagellation and gametogenesis in vitro, we subjected the samples to a Percoll Density Gradient to separate the parasites from the rest of the blood cells. Following centrifugation, the layer containing extracellular parasites underwent a flow cytometry and cell sorting process, during which we selected two different subpopulations of cells for analysis. Based on qPCR analyses, we demonstrate parasite DNA enrichment in Percoll Density Gradient and flow cytometry samples; simultaneously, these samples showed the lowest concentration of Columba livia DNA. However, the concentration of parasite DNA was higher in the PDG than in the cell sorting sample. This study reports the concentration of the Haemoproteus parasite by flow cytometry without DNA-intercalating dyes, and this methodology can serve as a technique for DNA enrichment of blood parasites infecting nucleated red blood cells to improve techniques that allow obtaining complete genomes.

Sediment sorting and transport mechanism controlled by both soil properties and hydraulic parameters on hillslopes

Understanding the sediment sorting and transport processes is is crucial for enhancing the understanding of soil erosion mechanisms and improving soil erosion prediction. However, quantitative studies based on the coupling effects of both hydraulic parameters and soil properties are still lacking. In this study, five loessial soils with varying physicochemical properties were collected in the Loess Plateau of China, and simulated rainfall experiments were carried out in a metal box with a slope of 20 degrees under rainfall intensities of 1, 1.5, and 2 mm min−1. Runoff samples were collected to analyze the variation in erosion rates and effective particle size distribution of eroded sediments during the rainfall process. The results showed that the stream power was the optimal parameter to model the erosion rate, and a power function incorporating the clay content and stream power could well predict the erosion rate of different soils. The enrichment ratio indicated the enrichment of fine particles and depletion of coarse particles. And the sediment sorting degree increased with increasing soil clay content. The critical particle size for distinguishing different transport mechanisms varied among soils, with more than 55.46 % of the total sediment transported by suspension/saltation. A linear function and power function including the original soil particle size distribution and stream power were developed to quantify the sediment clay content and the relative contribution of different transport mechanisms. This study could be helpful to improve the soil erosion prediction model and provide basic guidelines for regional soil and water conservation.

Hands-On Activity Illustrating the Sorting Process of Recycled Waste and Its Role in Promoting Sustainable Solutions.

Effective recycling is not merely a matter of collecting waste; it also requires meticulous categorization to maximize the potential for reusing material and minimizing waste sent to landfills. Education and awareness of the sorting and recycling process bottlenecks need to be emphasized and extended beyond higher educational contexts (e.g., in multiple stages of educational pathways, such as middle or high school). Hence, this project introduces a hands-on plastic sorting activity where students use recycled waste to be sorted based on their physical properties. Several tools were provided to perform the separation, such as water, sieves, magnets, and manual/visual separation while considering the time and cost associated with each tool. The activity was evaluated by pre- and post-evaluations based on Likert-scale and open-ended questions grouped in several categories related to the sorting process. In several categories, we observed that the activity enhanced student knowledge (e.g., general sorting understanding), while in other categories, there was no growth. From open-ended questions, students expressed an understanding of how to sort recycled waste and an appreciation for the trade-offs in developing sorting solutions. This activity effectively enhanced students' awareness of the sorting process of recycled waste. It lays the foundation for future inquiry and outreach project design.

Industrialized fine physical regeneration process and economic benefit assessment for recycling waste HDPE containers

The production of waste HDPE containers is enormous, and its recycling and utilization is an inevitable trend to achieve sustainable social development. However, the engineering practice of recycling waste HDPE containers is still lacking. This study provides a detailed explanation of the principles of fine recycling for waste HDPE containers, and develops a set of physical regeneration processes. By combining visual recognition assisted sorting, alkali hot water washing, physical sorting, and regranulation processes, high-quality regenerated HDPE particles are prepared. Meanwhile, various real data (sample size 50–100) of the production line and products are tracked, statistically analyzed. The results show that regenerated HDPE particles have good physical properties, the ratio of HDPE particles containing aluminum foil and multicolored impurities are extremely low, and the ratio of HDPE particles containing black spots (d＞0.1 mm) is less than 1.61 × 10−2 wt% (mean). Economic benefit analysis shows that the gross profit of recycled HDPE particles is USD $312.05 per ton. The above conclusions prove the feasibility and progressiveness of this process for the regeneration of waste HDPE containers. Meanwhile, this study provides engineering technical guidance for the large-scale fine recycling and regeneration of waste HDPE containers.

Do the institutionally disadvantaged students benefit more from boarding at school in their studying? The role of migration and Hukou status in China

While the impact of boarding school on academic achievement has been extensively debated, the differential effects on distinct student groups divided institutionally have yet to be thoroughly explored. Moreover, previous studies focus on comparing boarding schools vs. day schools, which tend to suffer from bias due to the sorting processes of different school types. In this study, we employ the school-fixed effect model and inverse probability weighting approach to examine the association of boarding on middle-school students’ academic outcomes in China, where most schools have a mixed population of boarders and non-boarders. Based on the two waves of panel data from the China Education Panel Survey (CEPS) in 2013 and 2014, we find that boarding is positively associated with students’ academic scores, particularly math and Chinese. Rural-origin students, including rural local students and rural migrant students, benefit more from living on campus. Reduced study distractions, such as housework and playing can partially explain this positive boarding effect. This study lends empirical support to the positive effect of boarding on academic performance among students who are segregated by Hukou and migration status in China.

Wave‐Driven Vertical Sorting of Density‐Varying Particles

AbstractIt has been recognized that the vertical sorting of polydispersed sand grains via the “Brazil Nut effect” can lead to inverse grading (upward coarsening) in the surface layer of a sand bed. However, the addition of nonnative particles not inherently observed on the beach, characterized by different densities and sizes, may complicate the vertical sorting processes and the fate of the nonnative particles. For example, olivine particles, which may be released on natural sandy beaches to facilitate carbon dioxide capture via weathering from wave action, have a density about 25% larger than the typical native sands. An Eulerian‐Lagrangian model, which couples Computational Fluid Dynamics for the fluid phase and Discrete Element Method for the particle phase was utilized to investigate how the so‐called “Reverse Brazil Nut effect” due to the larger density of nonnative (olivine) particles may be counteracted by the so‐called “Brazil Nut effect” for a range of nonnative particle sizes. Numerical simulations showed that the higher‐density nonnative particles tended to sink into the sand bed consistent with the “Reverse Brazil Nut effect”; however, the vertical sorting structure of native sand, driven by the “Brazil Nut effect,” also was a controlling factor determining the fate of nonnative particles. To maintain the presence of the nonnative (olivine) particles in the active layer of sediment transport, the higher‐density nonnative particle size must be larger than the mean native sand size found at the bottom of the active layer.

Optical detection of plastic waste through computer vision

An incredibly wide range of plastic products and enormous volumes of generated plastic wastes makes sorting technologies highly important. In order to enhance the efficiency and accuracy of sorting process, this article proposes an Internet of Video Things solution based on the deep learning algorithms for image recognition of plastic waste on a moving conveyor belt and embedded intelligence. In particular, the state-of-the-art object detection models, including Faster R-CNN, RetinaNet and YOLOv8 are used. Target categories of plastic are Polyethylene Terephthalate (PET) and Polypropylene (PP). Furthermore, we implemented quantization techniques for trained models on a commercial off-the-shelf embedded system for fast processing time. We achieved a high mean Average Precision (mAP) metric of 77.74% and accuracy of 95.67% on a test set and are fine-tuned and optimized for the deployment on an Nano embedded system providing 20 frames per second. This research contributes to the field of application of Internet of Things by demonstrating the efficacy of deep learning algorithms run on the embedded system in the industrial plastic waste sorting process. The findings highlight the practical applicability of these algorithms and offer insights into resource management and recycling practices.

Classification of hazelnut varieties based on bigtransfer deep learning model

Hazelnut is an agricultural product that contributes greatly to the economy of the countries where it is grown. The human factor plays a major role in hazelnut classification. The typical approach involves manual inspection of each sample by experts, a process that is both labor-intensive and time-consuming, and often suffers from limited sensitivity. The deep learning techniques are extremely important in the classification and detection of agricultural products. Deep learning has great potential in the agricultural sector. This technology can improve product quality, increase productivity, and offer farmers the ability to classify and detect their produce more effectively. This is important for sustainability and efficiency in the agricultural industry. In this paper aims to the application of deep learning algorithms to streamline hazelnut classification, reducing the need for manual labor, time, and cost in the sorting process. The study utilized hazelnut images from three different varieties: Giresun, Ordu, and Van, comprising a dataset of 1165 images for Giresun, 1324 for Ordu, and 1138 for Van hazelnuts. This dataset is an open-access dataset. In the study, experiments were carried out on the determination of hazelnut varieties with BigTransfer (BiT)-M R50 × 1, BiT-M R101 × 3 and BiT-M R152 × 4 models. Deep learning models, including big transfer was employed for classification. The classification task involved 3627 nut images and resulted in a remarkable accuracy of 99.49% with the BiT-M R152 × 4 model. These innovative methods can also lead to patentable products and devices in various industries, thereby boosting the economic value of the country.