Template-Type: ReDIF-Article 1.0 Author-Name: James Roy Lesidan Author-Workplace-Name: Department of Physics, College of Arts and Sciences, Visayas State University, Baybay City, Philippines Author-Name: Mencius Lesidan Author-Workplace-Name: National Coconut Research Center - Visayas, Visayas State University, Baybay City, Philippines Author-Name: Melvin Hagonob Author-Workplace-Name: Prairie Swine Research Inc., University of Saskatchewan, Saskatoon, Canada Author-Name: Charlie Andan Author-Workplace-Name: Department of Meteorology, College of Engineering and Technology, Visayas State University, Baybay City, Philippines Author-Workplace-Name: Institute of Chemical Processes Fundamentals of Czech Academy of Sciences, Prague, Czech Republic Author-Name: Ma. Grace Sumaria Author-Workplace-Name: Department of Agricultural and Biosystems Engineering, College of Engineering and Technology, Visayas State University, Baybay City, Philippines Author-Name: Ronaldo Almencion Author-Workplace-Name: Department of Agricultural and Biosystems Engineering, College of Engineering and Technology, Visayas State University, Baybay City, Philippines Author-Name: Kebin Ysrael Martinez Author-Workplace-Name: JE Hydro & Bio-Energy Corporation, Cebu City, Philippines Title: Portable analogue-based electronic moisture meter for root-crop chips Abstract: Moisture content regulation of root crops is crucial in post-harvest processing operations, not only in the price stipulation but also to avoid aflatoxin contamination. To prolong their storage life, they are processed into dried chips to extend their usability in feed formulations and starches. In this study, we use the capacitance-based method to evaluate the performance of an analogue-based electronic meter for the cassava, sweet potato, and taro chips. The meter was calibrated against the oven-drying method, yielding high R2 values of the different root crops. The established calibration models were validated and revealed high R2 values with 0.9580 for the cassava, 0.9958 for the sweet potato, and 0.9798 for the taro. The trendline equations are y = 59.44x0.56, y = 54.38x0.47, and y = 52.94x0.62, respectively. The results revealed that the moisture meter is capable of reading the moisture content on a weight basis (% MCwb) with accuracy and reliability at specified limits of 8% < x < 69% for the cassava, 15% < x < 59% for the sweet potato, and 9% < x < 57% for the taro. This study presents the performance of a portable analogue-based moisture meter as a reliable and accessible solution to small-scale operations, especially for farmers, offering an on-site rapid moisture content measurement in root crop processing. Keywords: capacitance-based, moisture content, root crop processing, dielectric property, regression analysis, post-harvest technology Journal: Research in Agricultural Engineering Pages: 113-120 Volume: 71 Issue: 2 Year: 2025 DOI: 10.17221/15/2025-RAE File-URL: http://rae.agriculturejournals.cz/doi/10.17221/15/2025-RAE.html File-Format: text/html X-File-Ref: http://agriculturejournals.cz/RePEc/caa/references/rae-202502-0001.txt Handle: RePEc:caa:jnlrae:v:71:y:2025:i:2:id:15-2025-RAE Template-Type: ReDIF-Article 1.0 Author-Name: Ninja Begum Author-Workplace-Name: Department of Food Engineering and Technology, Tezpur University, Assam, India Author-Name: Manuj Kumar Hazarika Author-Workplace-Name: Department of Food Engineering and Technology, Tezpur University, Assam, India Title: Spoilage detection of tomatoes using a convolutional neural network Abstract: With the increasing productivity in agriculture, it has become extremely essential to look for an advanced technique that will help to minimise losses. Recently, deep learning has outperformed the task of recognition and classification of fruits and vegetables automatically from images, finding applicability in this study. This work, thus, attempts to develop an automatic spoilage detection CNN model for tomatoes. In this work, a deep learning-based CNN model is trained and validated on a self-prepared dataset for classifying tomatoes as edible and spoilt is proposed. The dataset consisted of 810 images, out of which 572 images were considered for training and 238 images for validation. The model is also trained iteratively with varying epoch and batch sizes to evaluate the model in giving the highest classification accuracy. The highest accuracy of 99.70% was achieved at epoch 20 and batch size 32. Further evaluating the performance of the developed model using a confusion matrix, a precision, recall and accuracy of 100%, 87% and 95%, respectively, was obtained for the spoilage detection of tomatoes. Also, on establishing Pearson's correlation between the predictive model and the sensory evaluation results, a Pearson correlation of 0.895 was obtained, showing that there is strong linear correlation between them. Keywords: tomato samples, training datasets, classification models, models accuracy, images validation Journal: Research in Agricultural Engineering Pages: 80-87 Volume: 71 Issue: 2 Year: 2025 DOI: 10.17221/31/2024-RAE File-URL: http://rae.agriculturejournals.cz/doi/10.17221/31/2024-RAE.html File-Format: text/html X-File-Ref: http://agriculturejournals.cz/RePEc/caa/references/rae-202502-0002.txt Handle: RePEc:caa:jnlrae:v:71:y:2025:i:2:id:31-2024-RAE Template-Type: ReDIF-Article 1.0 Author-Name: Sitti Nur Faridah Author-Workplace-Name: Agricultural Technology Department, Faculty of Agriculture, Hasanuddin University, Makassar, Indonesia Author-Name: Muhammad Tahir Sapsal Author-Workplace-Name: Agricultural Technology Department, Faculty of Agriculture, Hasanuddin University, Makassar, Indonesia Author-Name: Tisha Aditya A. Jamaluddin Author-Workplace-Name: Energy Conversion and Conservation Research Centre, National Research and Innovation Agency,South Tangerang Banten, Indonesia Author-Name: Andini Dani Achmad Author-Workplace-Name: Electrical Engineering Department, Faculty of Engineering, Hasanuddin University, Gowa, Indonesia Author-Name: Muhammad Adi Surya Author-Workplace-Name: Agricultural Technology Department, Faculty of Agriculture, Hasanuddin University, Makassar, Indonesia Title: Stability of soil moisture sensors for agricultural crop cultivation Abstract: Soil water content is critical in plants' morphological and physiological processes; therefore, water must always be available in appropriate quantities to meet plant growth needs. Soil moisture can be easily detected using sensors, which offer a practical solution for monitoring water content in the soil. However, using sensors for a long time, especially on agricultural land, will reduce sensor accuracy. This research aims to investigate the accuracy of soil moisture sensors during their use for cultivating crops. Using sensors in sandy clay soil can detect soil moisture levels with an accuracy of 93.80% and a precision of 90.81%. A reading deviation (error) of up to 49.74% with a precision level of 75.69% occurred when the sensor had been used for 40 days. Regular cleaning and calibration of the sensor are necessary to obtain accurate soil moisture readings. A copper-based sensor module kit can be used to detect soil moisture with reasonable accuracy during plant growth with a 5-6 weeks harvest time. Keywords: agricultural land, copper sensor, sandy clay, soil water content Journal: Research in Agricultural Engineering Pages: 88-94 Volume: 71 Issue: 2 Year: 2025 DOI: 10.17221/33/2024-RAE File-URL: http://rae.agriculturejournals.cz/doi/10.17221/33/2024-RAE.html File-Format: text/html X-File-Ref: http://agriculturejournals.cz/RePEc/caa/references/rae-202502-0003.txt Handle: RePEc:caa:jnlrae:v:71:y:2025:i:2:id:33-2024-RAE Template-Type: ReDIF-Article 1.0 Author-Name: Jaroslav Mrázek Author-Workplace-Name: Department for Quality and Dependability of Machines, Faculty of Engineering, Czech University of Life Sciences Prague, Prague, Czech Republic Author-Name: Jakub Vošáhlík Author-Workplace-Name: Department of Technological Equipment of Constructions, Faculty of Engineering,Czech University of Life Sciences Prague, Prague, Czech Republic Author-Name: Eva Olmrová Author-Workplace-Name: Department for Quality and Dependability of Machines, Faculty of Engineering, Czech University of Life Sciences Prague, Prague, Czech Republic Author-Name: Martin Pexa Author-Workplace-Name: Department for Quality and Dependability of Machines, Faculty of Engineering, Czech University of Life Sciences Prague, Prague, Czech Republic Author-Name: Zdeněk Aleš Author-Workplace-Name: Department for Quality and Dependability of Machines, Faculty of Engineering, Czech University of Life Sciences Prague, Prague, Czech Republic Author-Name: Jakub Čedík Author-Workplace-Name: Department for Quality and Dependability of Machines, Faculty of Engineering, Czech University of Life Sciences Prague, Prague, Czech Republic Title: Camera systems and their user recognition reliability when entering an agri-food complex Abstract: This study evaluates the efficiency of various facial recognition camera systems used to control access in agri-food production environments, focusing on their ability to identify individuals based on biometric facial traits. It is also important to prevent the movement of unwanted persons into the production premises in the agri-food complex. The main goal was to assess how these factors influence the recognition performance and to determine the most reliable system for preventing unauthorised entry. The results show notable performance disparities between the devices tested. It can be concluded in this research that there are statistically significant differences between the maternal, professional and semi-professional systems. The device that is most suited is the HIKVISION iDS-2CD8426G0/F-I, achieving the best average performance score. This is based on usual recognition times. These tests indicate that the HIKVISION DS-2DE7232IW-AE(S5), which obtained an average rating of 2.216789, is the second-best acceptable device. With a score of 2.842113, HIKVISION DS-2CD2H45FWD-IZS (2.8-12 mm) (B) received, without a doubt, the lowest ranking. Given the outcomes, systems with superior recognition capabilities like the iDS-2CD8426G0/F-I are best to use for critical access control applications and to also minimise the use of facial coverings in sensitive areas to ensure reliable identification and higher levels of security of agri-food complexes. Keywords: security, agricultural buildings, ergonomics, facial recognition, face detection Journal: Research in Agricultural Engineering Pages: 105-112 Volume: 71 Issue: 2 Year: 2025 DOI: 10.17221/35/2025-RAE File-URL: http://rae.agriculturejournals.cz/doi/10.17221/35/2025-RAE.html File-Format: text/html X-File-Ref: http://agriculturejournals.cz/RePEc/caa/references/rae-202502-0004.txt Handle: RePEc:caa:jnlrae:v:71:y:2025:i:2:id:35-2025-RAE Template-Type: ReDIF-Article 1.0 Author-Name: Anita Chidera Ezeagba Author-Workplace-Name: Department of Biosystems Engineering, University of Manitoba, Winnipeg, Manitoba, Canada Author-Name: Cheryl Mary Glazebrook Author-Workplace-Name: Faculty of Kinesiology and Recreation Management, University of Manitoba, Winnipeg, Manitoba, Canada Author-Name: Daniel Delmar Mann Author-Workplace-Name: Department of Biosystems Engineering, University of Manitoba, Winnipeg, Manitoba, Canada Title: Perception of bimodal warning cues during remote supervision of autonomous agricultural machines Abstract: Agricultural machines that are fully autonomous will still need human supervisors to monitor and troubleshoot system failures. Recognising the emergency as soon as possible is crucial to reduce adverse effects. The ability of humans to detect visual, auditory, or tactile cues is usually enabled by warning systems. The effectiveness of different warning cues varies in terms of prompting a quick response. The study's objective was to compare the effectiveness of two bimodal warnings (i.e., visual-auditory and visual-tactile) at eliciting supervisor perception (which equates to level one situation awareness). Twenty-five participants engaged in an autonomous sprayer simulation. Two realistic remote supervision scenarios (i.e., in-field and close-to-field) were used to examine two bimodal warning cues: (i) visual-auditory and (ii) visual-tactile. The effectiveness of each bimodal warning was assessed based on two measures: (i) response time and (ii) noticeability. There was no significant difference between the bimodal warning cues in terms of response time when tractor sound was present in the experimental environment (reflecting the in-field remote supervision scenario); however, visual-tactile cues yielded shorter response times than visual-auditory cues when the experimental environment was quiet (reflecting the close-to-field remote supervision scenario). There were no statistically significant differences between visual-auditory and visual-tactile warnings concerning noticeability. Participants' subjective answers indicated they preferred the visual-tactile cues better than the visual-auditory cues. It is concluded that visual-tactile warnings are preferred over visual-auditory warnings to enable perception during remote supervision of autonomous agricultural machines (AAMs). Keywords: warning systems, situation awareness, human supervision, automated farm machinery Journal: Research in Agricultural Engineering Pages: 69-79 Volume: 71 Issue: 2 Year: 2025 DOI: 10.17221/73/2024-RAE File-URL: http://rae.agriculturejournals.cz/doi/10.17221/73/2024-RAE.html File-Format: text/html X-File-Ref: http://agriculturejournals.cz/RePEc/caa/references/rae-202502-0005.txt Handle: RePEc:caa:jnlrae:v:71:y:2025:i:2:id:73-2024-RAE Template-Type: ReDIF-Article 1.0 Author-Name: Ehsan Sheidaee Author-Workplace-Name: Biosystems Engineering Department, Tarbiat Modares University, Tehran, Iran Author-Name: Pourya Bazyar Author-Workplace-Name: Department of Mechanical Engineering and Production Management, Hamburg University of Applied Science, Hamburg, Germany Title: Enhancing the destructive egg quality assessment using the machine vision and feature extraction technique Abstract: The rapid growth of the food industry necessitates rigorous quality control, particularly in egg production. This study explores advanced methodologies for egg quality assessment by integrating the Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and k-Nearest Neighbour (KNN) with machine vision techniques. While traditional destructive methods like measuring the Haugh unit (HU) offer direct insights, but render eggs unusable, non-destructive techniques, such as imaging and spectroscopy, allow continuous quality monitoring. Over a 20-day period, egg samples were evaluated using a digital camera to capture key parameters like the albumen and yolk heights. The study's image processing involved noise reduction, feature extraction, and calibration. The PCA captured 90.18% of the data variability, while LDA achieved 100% classification accuracy, and KNN demonstrated 80% accuracy. These findings underscore the effectiveness of combining machine vision with statistical methods to enhance the egg grading accuracy, contributing to consumer safety and industry standards. Keywords: haugh unit, image processing, classification, quality control, albumin height Journal: Research in Agricultural Engineering Pages: 95-104 Volume: 71 Issue: 2 Year: 2025 DOI: 10.17221/86/2024-RAE File-URL: http://rae.agriculturejournals.cz/doi/10.17221/86/2024-RAE.html File-Format: text/html X-File-Ref: http://agriculturejournals.cz/RePEc/caa/references/rae-202502-0006.txt Handle: RePEc:caa:jnlrae:v:71:y:2025:i:2:id:86-2024-RAE