Search

With the rapid growth of China's economy and the supportive government policies and funds, the demand for and the expansion of rural housing in China are on the rise, which tightens requirements for rural housing. However, most of the present rural housing suffers from unplanned growth. The sprawl of rural housing adversely affects the quantity and quality of land resources, in particular, productive agricultural lands; therefore, it is necessary to regulate the growth of rural housing and protect farmlands through spatiotemporally continuous monitoring. Currently, the monitoring of rural housing in China is mainly conducted via in-situ inspection of the national land survey, which is restricted by unfavorable conditions (e.g., weather, outbreaks, and traffic) as well as impairing real-time and reliable control over information collected. To resolve this issue, this study proposed an intelligent model to recognize rural housing in underdeveloped areas based on deep learning and mathematical morphology (MobileNet-MM). The model was based on high-resolution remote sensing data, MobileNetV2 (a convolutional neural network architecture well performing on mobile devices), and mathematical morphology. First, the obtained data were segmented and manually screened and tagged to construct a training dataset. Second, the training dataset was used to train MobileNet-MM, with the expansion operation being used to compensate for identification errors of deep learning. Finally, the accuracy of MobileNet-MM to identify and monitor rural housing was tested, resulting in 84.5% accuracy. The comparison of the accuracies of MobileNet-MM and ResNet34 (a state-of-the-art image classification model) indicated that ResNet34 misclassified a large area of rural housing that was mainly distributed on the edge of the region as well as cropland and vegetation as rural housing, with its weak ability to recognize actual rural housing. The MobileNet-MM model predicted rural housing accurately and land boundary precisely, with the misclassified area being scattered, and its average accuracy, is 10.6% higher than that of ResNet34. The novelties of this study were two-fold: (1) a high-resolution training dataset of rural housing in underdeveloped areas was generated, which provides data support for the development of subsequent models; and (2) an intelligent model to recognize rural housing in underdeveloped areas (MobileNet-MM) based on deep learning and mathematical morphology was proposed. [ABSTRACT FROM AUTHOR]

Since tlie refonn and opening up, China has achieved great progress in industrialization and urbanization. However, due to tlie long-tenn implementation of the economic development model of "emphasizing cities over township" and tlie urban-rural dual system, the gap between urban and rural development is constantly increasing, and the problems of land resource allocation and contradiction between man and land in rural areas have gradually emerged. Rural homestead is the main place where fanners live, witli unique land use characteristics. The infbnuation such as tlie structure and utilization status of houses on it can clearly reflect the local rural economic situation and homestead problems. Therefore, understanding tlie characteristics and differences of rural homestead housing is of great significance to understand and grasp the situation and problems of rural development and promote the implementation of rural revitalization strategy. Based on the rural cadastral survey data of Guangdong Province and combined witli G1S spatial analysis method, this paper analyzes the characteristics of rural homestead housing. The results show that: (1) The rural housing in the study area is relatively low, mainly with one floor, supplemented by two and tliree floors, mainly brick and mixed structure, supplemented by concrete structure, and there is a certain correlation between tlie number of floors and the spatial distribution of the structure. Among them, the proportions of one storey house, brick structure and mixed structure house are 59.42%, 37.71% and 42.60% respectively. (2) There are obvious regional diflferences in tlie number of floors and structural spatial distribution of rural homestead housing in tlie study area. Low-rise, brick and mixed structure houses are mainly distributed in remote areas such as nortliem and western Guangdong, and high-rise and concrete structure houses are mostly concentrated in the Pearl River Delta. (3) The vacancy and idleness of rural homestead housing in tlie study area are common. The average vacancy rate and idle rate are 6.75% and 5.66% respectively, and tlie rate of one houseliould witli multiple houses reaches 23.88%. The phenomenon of rural hollowing is obvious, and 50.87% of the villages in districts and counties are above the moderate hollowing, mainly distributed in tlie fringe area of the Pearl River Delta and Meizhou in nortliem Guangdong. At the same time, tliere is a certain phenomenon that tlie average household homestead area exceeds tlie standard. (4) The characteristics of rural housing utilization in Guangdong Province are aflected by factors such as policies, economy, transportation, physical geographical condition and local customs. And the influencing factors in different regions are diflerent, such as tlie special folk customs have a greater impact on rural housing in eastern Guangdong, and the economic situations and urban siphon have a greater impact on the characteristics of rural housing in the edge of the Pearl River Delta. Tliere are obvious spatial differences in the utilization of rural homestead housing in Guangdong Province, and tliere are many idle homestead resources. There is great potential for tlie optimization and adjustment of rural construction land for rural revitalization. [ABSTRACT FROM AUTHOR]