RURAL SETTLEMENTS SEGMENTATION BASED ON DEEP LEARNING U-NET USING REMOTE SENSING IMAGES تقسيم المستوطنات الريفية على أساس التعلم العميق U - NET باستخدام صور الاستشعار عن بعد RURAL SETTLEMENTS SEGMENTATION BASED ON DEEP LEARNING U-NET USING REMOTE SENSING IMAGES SEGMENTATION DES RÈGLEMENTS RURAUX BASÉE SUR L'APPRENTISSAGE EN PROFONDEUR U-NET À L'AIDE D'IMAGES DE DÉTECTION À DISTANCE RURAL SETTLEMENTS SEGMENTATION BASED ON DEEP LEARNING U-NET USING REMOTE SENSING IMAGES

Abstract. Accurate and efficient extraction of rural settlements from high-resolution remote sensing imagery is of paramount importance for rural government management. Unplanned rural settlements are quite common. Understanding the spatial characteristic of these rural settlements is of great importance as it offers indispensable information for land management and decision-making. In this setting, the U-net architecture is proposed in this study for rural settlements differentiation by image segmentation on high-resolution satellite images of rural settlements in Zagora province, Draa-Tafilalet region, Morocco. To predict pixels in remote sensing images representing rural settlements in this province. Image segmentation is conducted using different encoders in the U-net architecture, and the results are compared. Experimental results demonstrate that the proposed method effectively mapped and discriminated rural settlements areas with an overall accuracy of 98%, achieving comparable and improved performance over other traditional rural extraction methods. الخلاصة: إن الاستخراج الدقيق والفعال للمستوطنات الريفية من صور الاستشعار عن بعد عالية الدقة له أهمية قصوى لإدارة الحكومة الريفية. المستوطنات الريفية غير المخططة شائعة جدًا. إن فهم السمة المكانية لهذه المستوطنات الريفية له أهمية كبيرة لأنه يوفر معلومات لا غنى عنها لإدارة الأراضي وصنع القرار. في هذا السياق، تم اقتراح بنية U - net في هذه الدراسة لتمييز المستوطنات الريفية عن طريق تجزئة الصور على صور الأقمار الصناعية عالية الدقة للمستوطنات الريفية في مقاطعة زاغورا، منطقة درعة تافيلالت، المغرب. التنبؤ بالبكسل في صور الاستشعار عن بعد التي تمثل المستوطنات الريفية في هذه المقاطعة. يتم إجراء تجزئة الصور باستخدام أجهزة تشفير مختلفة في بنية U - net، وتتم مقارنة النتائج. تُظهر النتائج التجريبية أن الطريقة المقترحة رسمت خرائط فعالة ومميزة لمناطق المستوطنات الريفية بدقة إجمالية بلغت 98 ٪، وحققت أداءً قابلاً للمقارنة ومحسّنًا مقارنة بطرق الاستخراج الريفية التقليدية الأخرى. Abstract. Accurate and efficient extraction of rural settlements from high-resolution remote sensing imagery is of paramount importance for rural government management. Unplanned rural settlements are quite common. Understanding the spatial characteristic of these rural settlements is of great importance as it offers indispensable information for land management and decision-making. In this setting, the U-net architecture is proposed in this study for rural settlements differentiation by image segmentation on high-resolution satellite images of rural settlements in Zagora province, Draa-Tafilalet region, Morocco. To predict pixels in remote sensing images representing rural settlements in this province. Image segmentation is conducted using different encoders in the U-net architecture, and the results are compared. Experimental results demonstrate that the proposed method effectively mapped and discriminated rural settlements areas with an overall accuracy of 98%, achieving comparable and improved performance over other traditional rural extraction methods. Abstract. Accurate and efficient extraction of rural settlements from high-resolution remote sensing imagery is of paramount importance for rural government management. Les règlements ruraux non planifiés sont communs. Comprendre la caractéristique spatiale de ces règlements ruraux est d'une grande importance car il offre des informations essentielles pour la gestion des terres et la prise de décision. In this setting, the U-net architecture is proposed in this study for rural settlements differentiation by image segmentation on high-resolution satellite images of rural settlements in Zagora province, Draa-Tafilalet region, Morocco. To predict pixels in remote sensing images representing rural settlements in this province. La segmentation d'image est conduite en utilisant des encodeurs différents dans l'architecture U-net, et les résultats sont comparés. Experimental results demonstrate that the proposed method effectively mapped and discriminated rural settlements areas with an overall accuracy of 98%, achieving comparable and improved performance over other traditional rural extraction methods. Abstract. Accurate and efficient extraction of rural settlements from high-resolution remote sensing imagery is of paramount importance for rural government management. Unplanned rural settlements are quite common. Understanding the spatial characteristic of these rural settlements is of great importance as it offers indispensable information for land management and decision-making. In this setting, the U-net architecture is proposed in this study for rural settlements differentiation by image segmentation on high-resolution satellite images of rural settlements in Zagora province, Draa-Tafilalet region, Morocco. To predict pixels in remote sensing images representing rural settlements in this province. La segmentación de la imagen se realiza utilizando diferentes codificadores en la arquitectura U-net, y los resultados son comparados. Experimental results demonstrate that the proposed method effectively mapped and discriminated rural settlements areas with an overall accuracy of 98%, achieving comparable and improved performance over other traditional rural extraction methods.