阅读说明：本技术 一种山区土地面积测量方法及系统 (Mountain land area measuring method and system ) 是由 张斌 张晓静 赵丙辰 于 2019-12-09 设计创作，主要内容包括：本发明用于测量山区耕地或其他交通不便难以实施测量的土地面积,涉及北斗(或GPS)高精度定位的面积测量和基于遥感图像的语义分割算法等技术。方法步骤是：S1,沿环山公路使用北斗(或GPS)定位接收机测量待测土地所在的山区面积M,并在遥感地图上标注测量山区的封闭图形；S2,将山区封闭图形的遥感图像通过无线网络发送到服务器端程序,在服务器端使用深度神经网络的方法对该图像进行像素级语义分割,计算分割结果中待测土地图像特征像素点的数量g占整个封闭图形的像素点数量m的比值g/m。S3,将S1中的山地面积M与S2中得到的比值g/m相乘即得到山区中的耕地面积G。(The invention is used for measuring the land area of mountain cultivated land or other land which is inconvenient to traffic and difficult to measure, and relates to the technology of area measurement of Beidou (or GPS) high-precision positioning, semantic segmentation algorithm based on remote sensing images and the like. The method comprises the following steps: s1, measuring the area M of the mountain area where the land to be measured is located by using a Beidou (or GPS) positioning receiver along the mountain road, and marking a closed graph of the measured mountain area on a remote sensing map; s2, sending the remote sensing image of the mountain area closed graph to a server program through a wireless network, performing pixel-level semantic segmentation on the image at the server by using a deep neural network method, and calculating the ratio g/m of the number g of characteristic pixel points of the land image to be detected in a segmentation result to the number m of pixel points of the whole closed graph. And S3, multiplying the mountain land area M in the S1 by the ratio G/M obtained in the S2 to obtain the cultivated land area G in the mountain area.)

1. A method for measuring the land area in mountainous areas comprises the technologies of carrying out positioning sampling by a Beidou or GPS high-precision positioning receiver, forming a closed space, calculating the area of any closed graph, carrying out a remote sensing image semantic segmentation algorithm and the like, and is characterized in that: s1, measuring the area M of the mountain area where the land to be measured is located by using Beidou or GPS positioning along the mountain road; s2, generating a remote sensing image of a closed graph of the whole mountain, performing pixel-level semantic segmentation on the image by using a deep neural network method, and calculating the proportion g/m of the number g of characteristic pixel points of the image of the intertillage of a segmentation result to the number m of pixel points of the whole closed graph; and S3, multiplying the mountain land area M in the S1 by the proportion value obtained in the S2 by G/M to obtain the cultivated land area G in the mountain area.

2. The system of the method as claimed in claim 1, wherein Beidou or GPS high-precision positioning receiver is used for sampling and positioning the mountain location coordinates of the land to be measured, the mountain location coordinates are sent to the APP program of the smart phone end through a Bluetooth or 2.4G short-distance wireless communication method, the mountain area M of the land to be measured is marked and calculated on the remote sensing map by the APP program of the smart phone end, the remote sensing image marked with the closed graph is sent to the server end program through 4G or other wireless networks, the remote sensing image is subjected to image segmentation at the server end, the pixel point ratio G/M of the land to be measured is calculated, the land area G to be measured is calculated, and the calculated land area G is sent to the mobile phone end for display and output.

3. The system of the method as claimed in claim 1, wherein the mountain location coordinates of the land to be measured are located through Beidou or GPS sampling of a mobile phone, the mountain area M of the land to be measured is marked and calculated on a remote sensing map in an APP program, the remote sensing image marked with a closed graph is sent to a server end program through a 4G or other wireless network, the remote sensing image is subjected to image segmentation at the server end, the pixel point occupation ratio G/M of the land to be measured is calculated, the land area G to be measured is calculated, and the image is sent to the mobile phone end for display and output.

4. The system of the method as claimed in claim 1, wherein a Beidou or GPS high-precision positioning receiver is used for sampling and positioning the position coordinates of the mountainous area where the land to be detected is located, the coordinates are sent to a server end through a 4G or other wireless network, the mountainous area M where the land to be detected is located is marked and calculated on a remote sensing map in a server end program, then the remote sensing image marked with a closed graph is subjected to image segmentation, the pixel point occupation ratio G/M of the land to be detected is calculated, the land area G to be detected is calculated and output at the server end.

Technical Field

The invention is used for measuring the cultivated land in mountainous areas or agricultural land and other map areas which are inconvenient to carry out measurement and are difficult to carry out, relates to the technologies of area measurement of Beidou (or GPS) high-precision positioning, semantic segmentation algorithm based on remote sensing images and the like, and also relates to the development technology of a mobile phone APP terminal program and a server terminal program.

Background

The existing land and cultivated land area surveying and mapping method mainly comprises the following steps: 1. based on a tape measure or an infrared ruler, the method is easy to implement, but a large amount of manpower and material resources are consumed when the measurement area is large, and in addition, the measurement in the areas with inconvenient traffic in mountainous areas is dangerous; 2. the method is convenient and simple, but generally depends on the accuracy of the drawing, and needs a large amount of operation; 3. the method is measured by a total station area measuring method, has the highest precision, but needs a large amount of equipment and is difficult to implement in areas with inconvenient traffic; 4. the method and system for measuring land area under the application number of CN201910060718.2 provides a land area measuring method based on a mobile phone GPS (the contents of the above patents are introduced from the Chinese public network), which is simple and easy to implement, but the measuring precision is difficult to guarantee due to the limitation of the positioning precision of the GPS, and the measuring difficulty is increased due to the inconvenience of walking on mountain roads when measuring the cultivated land in mountain areas. 5. The method for estimating the green space coverage rate based on the remote sensing technology, which is applied under the application number of CN201611207011.2, provides an area estimation method through remote sensing image segmentation (the content of the above patent is introduced from the national knowledge network), but the method can only estimate the area due to the limitations of the resolution and the segmentation algorithm of the remote sensing image.

Disclosure of Invention

The invention is used for measuring the map and land area of mountain cultivated land or other traffic-inconvenient maps and land areas which are difficult to measure. The method is shown in figure 1, and comprises the following specific steps: s1, using big dipper (or GPS) to measure the area M of the mountain area where the land to be measured is located (at this time, M should completely contain the land G to be measured) along the mountain road, and marking the closed graph of the measured mountain area on the remote sensing map, where the specific method is shown in fig. 2, and the specific description is: s1.1, carrying out fixed-point sampling positioning along a highway under a mountain foot by using a Beidou (or GPS) high-precision positioning receiver, and sending the fixed-point sampling positioning to a mobile phone APP or a computer server side program, wherein the sampling points are selected at the turning positions of the mountain road, the more the sampling points are, the higher the measurement precision is, but the number of the sampling points is not more than 30; s1.2, marking on a remote sensing map after receiving coordinates of sampling points by a mobile phone APP end or a server end program (the remote sensing map of the mobile phone or the server program can be called by an API (application program interface) of a Baidu map or a Gaode map), automatically connecting the sampling points with the last sampling point on the remote sensing map in a straight line, and only making marking points on the remote sensing map if the sampling points are the first sampling points and not connecting the marking points; s1.3, when the last sampling point is collected, automatically connecting the point coordinate in the remote sensing map with the first sampling point coordinate by a program to form a closed graph, and calculating the area M of the whole closed graph, namely the projection area of the land to be detected sitting on the whole land mountain. S2, sending the marked remote sensing image of the mountain area closed area to a server-side program, performing pixel-level semantic segmentation on the image by using a deep neural network method at the server side, and calculating the proportion g/m of the number g of characteristic pixel points of the image of the intertillage map in the segmentation result to the number m of pixel points of the whole closed graph. And S3, multiplying the mountain land area M in the S1 by the proportion value obtained in the S2 by G/M to obtain the cultivated land area G in the mountain area.

Drawings

FIG. 1 is a flow chart of the overall steps of the system design of the present invention.

Fig. 2 is a flow chart of programming for collecting and calculating the area of a mountain area closed area at the APP end of the mobile phone.

Fig. 3 is a first embodiment of a system, a hardware solution layout. Wherein mark 1 is big dipper (or GPS) high accuracy positioning receiver, mark 2 is smart mobile phone end APP procedure, mark 3 is server end procedure, mark 4 is the bluetooth communication link between big dipper (or GPS) high accuracy positioning receiver and the smart mobile phone, mark 5 is the 4G network communication link between smart mobile phone and the server, mark 6 is big dipper (or GPS) location satellite, mark 7 is the remote sensing satellite.

FIG. 4 is a system embodiment layout. Wherein label 1 is big dipper (or GPS) high accuracy positioning receiver, and label 2 is the program of operation in cell-phone APP end, and label 3 is the program of operation in the server end.

FIG. 5 is a layout view of a second embodiment of the present invention. Wherein, label 1 is the smart phone APP terminal program, label 2 is the server terminal program, label 3 is the 4G network communication link between the smart phone and the server, label 4 big dipper (or GPS) positioning satellite, label 5 is the remote sensing satellite.

FIG. 6 is a design view of a third embodiment of the present invention. Wherein mark 1 is big dipper (or GPS) high accuracy positioning receiver, and mark 2 is server end procedure, and mark 3 is the 4G network communication link between big dipper (or GPS) high accuracy positioning receiver and the server, marks 4 big dipper (or GPS) positioning satellite, and mark 5 is the remote sensing satellite.

Detailed Description

Fig. 3 is an embodiment of the system of the present invention, and the specific embodiment is shown in fig. 4. The system consists of a high-precision positioning receiver, a mobile phone and a server (the high-precision positioning receiver comprises a Beidou high-precision positioning receiver, the GPS, Galileo and GLONASS positioning receivers are all within the protection range of the system, and the CORS base station can be used for providing high-precision positioning below the decimeter level), wherein the high-precision positioning receiver is used for carrying out high-precision positioning on a sampling point and sending to a mobile phone terminal APP program through Bluetooth communication (the communication scheme comprises but is not limited to Bluetooth communication, and the communication modes such as 2.4G, 4G, NFC and the like are all within the protection range of the system). The method comprises the steps that an APP program at a mobile phone end sequentially samples and marks a closed area on a map of a remote sensing satellite by using a sampling point positioning method, the terrain area is calculated, then mode selection can be carried out, if the model is a plain mode, the measured land area M is directly output, a mountain land mode is selected, a remote sensing satellite picture of the closed area is sent to a server end program through a 4G network, semantic segmentation of the remote sensing image is carried out, the proportion G/M of cultivated land pixel points in the picture is calculated, meanwhile, cultivated land area is obtained through M G/M calculation, and then the cultivated land area is sent to a mobile phone end through the 4G network to display the land area.

Fig. 5 is a second embodiment of the present invention, which directly uses the Beidou/GPS positioning and sampling of the mobile phone without using a high-precision positioning receiver, and the other mobile phone end programs are the same as the scheme shown in fig. 4.

Fig. 6 is a third embodiment of the present invention, in which a high-precision positioning receiver is used to directly communicate with a server through a 4G network, a server-side program directly marks a positioning point on a remote sensing map and forms a closed region, and then image segmentation and area calculation are performed, which has the same specific implementation as that shown in fig. 4, but all the programs are executed at the server side.