<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">vstisp</journal-id><journal-title-group><journal-title xml:lang="ru">Садоводство и виноградарство</journal-title><trans-title-group xml:lang="en"><trans-title>Horticulture and viticulture</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">0235-2591</issn><issn pub-type="epub">2618-9003</issn><publisher><publisher-name>Autonomous non-profit organization Editorial Board of journal «Horticulture and viticulture»</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.31676/0235-2591-2023-5-56-63</article-id><article-id custom-type="edn" pub-id-type="custom">jzxrso</article-id><article-id custom-type="elpub" pub-id-type="custom">vstisp-1091</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>ИНФОРМАЦИОННЫЕ ТЕХНОЛОГИИ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>INFORMATION TECHNOLOGY</subject></subj-group></article-categories><title-group><article-title>Сравнительный анализ моделей нейронных сетей для распознавания плодов яблони на кроне дерева</article-title><trans-title-group xml:lang="en"><trans-title>Neural network models of apple fruit identification in tree crowns: comparative analysis</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-7643-775X</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Кутырёв</surname><given-names>А. И.</given-names></name><name name-style="western" xml:lang="en"><surname>Kutyrev</surname><given-names>A. I.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Алексей Игоревич Кутырёв − кандидат технических наук, старший научный сотрудник</p><p>1-й Институтский проезд, д. 5, Москва, 109428</p></bio><bio xml:lang="en"><p>Alexey I. Kutyrev, PhD (Tech.), Senior Researcher</p><p>1st Institute Passage, 5, Moscow, 109428, Russia</p></bio><email xlink:type="simple">alexeykutyrev@gmail.com</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-9992-1261</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Смирнов</surname><given-names>И. Г.</given-names></name><name name-style="western" xml:lang="en"><surname>Smirnov</surname><given-names>I. G.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Доктор технических наук, главный научный сотрудник </p><p>1-й Институтский проезд, д. 5, Москва, 109428</p></bio><bio xml:lang="en"><p>Dr. Sci. (Tech.), Chief Researcher, Senior Researcher</p><p>1st Institute Passage, 5, Moscow, 109428, Russia</p></bio><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-0735-7697</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Андриянов</surname><given-names>Н. А.</given-names></name><name name-style="western" xml:lang="en"><surname>Andriyanov</surname><given-names>N. A.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Кандидат технических наук, специалист</p><p>1-й Институтский проезд, д. 5, Москва, 109428</p></bio><bio xml:lang="en"><p>PhD (Tech.), specialist</p><p>1st Institute Passage, 5, Moscow, 109428, Russia</p></bio><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Федеральный научный агроинженерный центр ВИМ</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Federal Scientific Agroengineering Center VIM</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2023</year></pub-date><pub-date pub-type="epub"><day>30</day><month>11</month><year>2023</year></pub-date><volume>0</volume><issue>5</issue><fpage>56</fpage><lpage>63</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Autonomous non-profit organization Editorial Board of journal «Horticulture and viticulture», 2023</copyright-statement><copyright-year>2023</copyright-year><copyright-holder xml:lang="ru">Autonomous non-profit organization Editorial Board of journal «Horticulture and viticulture»</copyright-holder><copyright-holder xml:lang="en">Autonomous non-profit organization Editorial Board of journal «Horticulture and viticulture»</copyright-holder><license xlink:href="https://www.sadivin.com/jour/about/submissions#copyrightNotice" xlink:type="simple"><license-p>https://www.sadivin.com/jour/about/submissions#copyrightNotice</license-p></license></permissions><self-uri xlink:href="https://www.sadivin.com/jour/article/view/1091">https://www.sadivin.com/jour/article/view/1091</self-uri><abstract><p>В статье представлены результаты проведённого с 2022 по 2023 гг. анализа качества работы современных моделей нейронных сетей для идентификации плодов яблони на кроне деревьев в изображениях. Для проведения исследований по выявлению наиболее качественного детектора использованы нейронные сети SSD (Single Shot MultiBox Detector), YOLOv4 (You Only Look Once версия 4), YOLOv5, YOLOv7 и YOLOv8. Оценивая качество работы рассматриваемых моделей для идентификации плодов, использовали метрики бинарной классификации Precision (точность), Recall (полнота), Accuracy (доля верных распознаваний), F-score (F-мера), AUC-ROCTotal (Area Under the Curve). Чтобы оценить точность прогноза идентификации плодов проведен расчет средней абсолютной ошибки (Mean Absolute Percentage Error, MAPE) анализируемых моделей нейронных сетей. Для анализа качества работы нейронной сети использовано 300 фотографий, сделанных в саду яблони. Анализ проведённых исследований показал, что модель SSD даёт более низкую скорость и точность работы, имеет высокие требования к вычислительным ресурсам, что может ограничить ее использование на менее производительных устройствах. Модель YOLOv4 превосходит YOLOv5 по точности на 10,2 %, однако при этом скорость обработки модели YOLOv5 превышает скорость YOLOv4 более чем в два раза. Это делает модель YOLOv5 более предпочтительным выбором для решения задач, связанных с обработкой больших объемов данных в режиме реального времени. Модель YOLOv8 превосходит модель YOLOv7 по скорости работы на 37,3 %, но точность у YOLOv7 выше на 9,4 %. Установлено, что наибольшее значение площади под Precision-Recall кривой составило 0,94 при использовании нейронной сети модели YOLOv7. Это указывает на высокую вероятность того, что классификатор сможет точно различить положительные и отрицательные значения класса плодов яблони. Проведенный расчет средней абсолютной ошибки (Mean Absolute Percentage Error, MAPE) анализируемых моделей нейронных сетей показал, что наименьшая ошибка идентификации плодов составила 5,64 % на модели YOLOv7 в сравнении с истинным значением, определенным визуальным методом. Анализ качества работы современных моделей нейронных сетей показал: семейство нейронных сетей You Only Look Once обеспечивает высокую скорость и точность обнаружения объектов, это позволяет им работать в режиме реального времени. Использование метода трансферного обучения (настройка только последних слоёв под решение узкоспециализированных задач) для адаптации работы моделей на различных сортах яблони позволит дополнительно повысить точность идентификации плодов.</p></abstract><trans-abstract xml:lang="en"><p>The article presents the results of an analysis conducted from 2022 to 2023 to assess the quality of modern neural network models of apple fruit identification in tree crowns shown in images. In order to conduct the studies on identifying the best detector, the following neural networks were used: SSD (Single Shot MultiBox Detector), YOLOv4 (You Only Look Once, Version 4), YOLOv5, YOLOv7, and YOLOv8. The performance of the considered models of apple fruit identification was assessed using such binary classification metrics as precision, recall, accuracy, F-score, and AUC-ROCTotal (area under the curve). To assess the accuracy in predicting apple fruit identification, the mean absolute percentage error (MAPE) of the analyzed neural network models was calculated. The neural network performance analysis used 300 photographs taken at an apple garden. The conducted studies revealed that the SSD model provides lower speed and accuracy, as well as having high requirements for computing resources, which may limit its use in lower performance devices. The YOLOv4 model surpasses the YOLOv5 model in terms of accuracy by 10.2 %, yet the processing speed of the YOLOv5 model is over twice that of the YOLOv4 model. This fact makes the YOLOv5 model preferable for tasks related to real-time big data processing. The YOLOv8 model is superior to the YOLOv7 model in terms of speed (by 37.3 %); however, the accuracy of the YOLOv7 model is 9.4 % higher. The highest area under the Precision-Recall curve amounts to 0.94 when using the YOLOv7 model. This fact suggests a high probability that the classifier can accurately distinguish between the positive and negative values of the apple fruit class. MAPE calculation for the analyzed neural network models showed that the lowest error in apple fruit identification amounted to 5.64 % for the YOLOv7 model as compared to the true value determined using the visual method. The performance analysis of modern neural network models shows that the YOLO family of neural networks provides high speed and accuracy of object detection, which allows them to operate in real time. The use of transfer learning (tuning of only the last layers to solve highly specialized problems) to adjust the performance of models for different apple fruit varieties can further improve the accuracy of apple fruit identification.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>сверхточная нейронная сеть</kwd><kwd>идентификация плодов яблони</kwd><kwd>машинное обучение</kwd><kwd>архитектура нейронной сети</kwd><kwd>компьютерное зрение</kwd><kwd>дистанционный мониторинг</kwd></kwd-group><kwd-group xml:lang="en"><kwd>high-accuracy neural network</kwd><kwd>apple fruit identification</kwd><kwd>machine learning</kwd><kwd>neural network architecture</kwd><kwd>computer vision</kwd><kwd>remote monitoring</kwd></kwd-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Ustin S. L., Gamon J. A. Remote sensing of plant functional types. New Phytologis. 2010;186(4):795-816.</mixed-citation><mixed-citation xml:lang="en">Ustin S. L., Gamon J. A. Remote sensing of plant functional types. New Phytologis. 2010;186(4):795-816.</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Penuelas J., Filella I., Gamon, J. A. Assessment of vegetation photosynthetic efficiency from space: A review. Remote Sensing of Environment. 2019;221:18-30.</mixed-citation><mixed-citation xml:lang="en">Penuelas J., Filella I., Gamon, J. A. Assessment of vegetation photosynthetic efficiency from space: A review. Remote Sensing of Environment. 2019;221:18-30.</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Singh A., Harrison P. Remote sensing of vegetation phenology. Remote Sensing of Environment. 2018;219:298-308.</mixed-citation><mixed-citation xml:lang="en">Singh A., Harrison P. Remote sensing of vegetation phenology. Remote Sensing of Environment. 2018;219:298-308.</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Shurygin B., Smirnov I., Chilikin A., Khort D., Kutyrev A., Zhukovskaya S., Solovchenko A. Mutual Augmentation of Spectral Sensing and Machine Learning for Non-Invasive Detection of Apple Fruit Damages. Horticulturae. 2022;8:1111.</mixed-citation><mixed-citation xml:lang="en">Shurygin B., Smirnov I., Chilikin A., Khort D., Kutyrev A., Zhukovskaya S., Solovchenko A. Mutual Augmentation of Spectral Sensing and Machine Learning for Non-Invasive Detection of Apple Fruit Damages. Horticulturae. 2022;8:1111.</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Yang J., Zhang H., Li Y. Application of Artificial Neural Networks in Fruit Classification. Journal of Physics: Conference Series. 2017;835(1):012044.</mixed-citation><mixed-citation xml:lang="en">Yang J., Zhang H., Li Y. Application of Artificial Neural Networks in Fruit Classification. Journal of Physics: Conference Series. 2017;835(1):012044.</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Wang X., Zhang Y., Ma Y. A Neural Network Approach to Fruit Detection and Recognition in Orchard Monitoring. IEEE Access. 2019;7:10618-10628.</mixed-citation><mixed-citation xml:lang="en">Wang X., Zhang Y., Ma Y. A Neural Network Approach to Fruit Detection and Recognition in Orchard Monitoring. IEEE Access. 2019;7:10618-10628.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Smirnov I., Kutyrev A., Kiktev N. Neural network for identifying apple fruits on the crown of a tree. E3S Web of Conferences, WFCES. 2021, 01021.</mixed-citation><mixed-citation xml:lang="en">Smirnov I., Kutyrev A., Kiktev N. Neural network for identifying apple fruits on the crown of a tree. E3S Web of Conferences, WFCES. 2021, 01021.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Zhao J., Wang Q., Cheng M., Wang W. Intelligent Detection of Fruit Defects Based on Deep Learning. Sensors. 2019;19(17):3679.</mixed-citation><mixed-citation xml:lang="en">Zhao J., Wang Q., Cheng M., Wang W. Intelligent Detection of Fruit Defects Based on Deep Learning. Sensors. 2019;19(17):3679.</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Ding H., Liu S., Fan X. Fruit detection using convolutional neural networks and deep learning. Computers and Electronics in Agriculture. 2019;157:453-460.</mixed-citation><mixed-citation xml:lang="en">Ding H., Liu S., Fan X. Fruit detection using convolutional neural networks and deep learning. Computers and Electronics in Agriculture. 2019;157:453-460.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Kandpal M., Jain P. K. Fruit quality assessment using artificial neural networks and machine vision. Journal of Food Science and Technology. 2020;57(8):2861-2873.</mixed-citation><mixed-citation xml:lang="en">Kandpal M., Jain P. K. Fruit quality assessment using artificial neural networks and machine vision. Journal of Food Science and Technology. 2020;57(8):2861-2873.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Zhang Y., Xu H., Fu W. A machine vision system based on artificial neural network for fruit detection and segmentation. Computers and Electronics in Agriculture. 2018;154:222-229.</mixed-citation><mixed-citation xml:lang="en">Zhang Y., Xu H., Fu W. A machine vision system based on artificial neural network for fruit detection and segmentation. Computers and Electronics in Agriculture. 2018;154:222-229.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Rahman M. A., Bala S. K. Fruit classification based on shape features using artificial neural network. Cogent Engineering. 2019;6(1):1692384.</mixed-citation><mixed-citation xml:lang="en">Rahman M. A., Bala S. K. Fruit classification based on shape features using artificial neural network. Cogent Engineering. 2019;6(1):1692384.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Kuznetsova A., Maleva T., Soloviev V. Using YOLOv3 Algorithm with Pre- and Post-Processing for Apple Detection in Fruit-Harvesting Robot. Agronomy. 2020;10:1016.</mixed-citation><mixed-citation xml:lang="en">Kuznetsova A., Maleva T., Soloviev V. Using YOLOv3 Algorithm with Pre- and Post-Processing for Apple Detection in Fruit-Harvesting Robot. Agronomy. 2020;10:1016.</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Andriyanov N., Khasanshin I., Utkin D., Gataullin T., Ignar S., Shumaev V., Soloviev V. Intelligent System for Estimation of the Spatial Position of Apples Based on YOLOv3 and Real Sense Depth Camera D415. Symmetry. 2022;14:148.</mixed-citation><mixed-citation xml:lang="en">Andriyanov N., Khasanshin I., Utkin D., Gataullin T., Ignar S., Shumaev V., Soloviev V. Intelligent System for Estimation of the Spatial Position of Apples Based on YOLOv3 and Real Sense Depth Camera D415. Symmetry. 2022;14:148.</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Milioto A., Lottes P., Stachniss C., Behnke S. Apple Detection in Orchards using Deep Convolutional Neural Networks. Sensors. 2018;18(6):1803.</mixed-citation><mixed-citation xml:lang="en">Milioto A., Lottes P., Stachniss C., Behnke S. Apple Detection in Orchards using Deep Convolutional Neural Networks. Sensors. 2018;18(6):1803.</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Wang Y., Zhang K., Zhang T., Song Z., Liu Y. Fruit Detection in Orchards using YOLOv4 Object Detection. IEEE Access. 2021;9:55828-55838.</mixed-citation><mixed-citation xml:lang="en">Wang Y., Zhang K., Zhang T., Song Z., Liu Y. Fruit Detection in Orchards using YOLOv4 Object Detection. IEEE Access. 2021;9:55828-55838.</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">Kandori T., Takimoto T., Hattori Y., Ohno K. Real-Time Fruit Detection System for Mobile Robot using YOLOv5. In 2021 16th International Conference on Computer Science &amp; Education (ICCSE). 2021, 56-61.</mixed-citation><mixed-citation xml:lang="en">Kandori T., Takimoto T., Hattori Y., Ohno K. Real-Time Fruit Detection System for Mobile Robot using YOLOv5. In 2021 16th International Conference on Computer Science &amp; Education (ICCSE). 2021, 56-61.</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">Yuan W. Accuracy Comparison of YOLOv7 and YOLOv4 Regarding Image Annotation Quality for Apple Flower Bud Classification. AgriEngineering. 2023;5:413-424.</mixed-citation><mixed-citation xml:lang="en">Yuan W. Accuracy Comparison of YOLOv7 and YOLOv4 Regarding Image Annotation Quality for Apple Flower Bud Classification. AgriEngineering. 2023;5:413-424.</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">Maxwell A., Warner T., Guillén L. Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies. Part 1: Literature Review. Remote Sens. 2021;13:2450.</mixed-citation><mixed-citation xml:lang="en">Maxwell A., Warner T., Guillén L. Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies. Part 1: Literature Review. Remote Sens. 2021;13:2450.</mixed-citation></citation-alternatives></ref><ref id="cit20"><label>20</label><citation-alternatives><mixed-citation xml:lang="ru">Kutyrev A., Kiktev N., Jewiarz M., Khort D., Smirnov I., Zubina V., Hutsol T., Tomasik M., Biliuk M. Robotic Platform for Horticulture: Assessment Methodology and Increasing the Level of Autonomy. Sensors. 2022;22:8901.</mixed-citation><mixed-citation xml:lang="en">Kutyrev A., Kiktev N., Jewiarz M., Khort D., Smirnov I., Zubina V., Hutsol T., Tomasik M., Biliuk M. Robotic Platform for Horticulture: Assessment Methodology and Increasing the Level of Autonomy. Sensors. 2022;22:8901.</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
