ArcGIS Tutorial: Overview of vectorization Techniques for ArcGIS Geology maps

Take ArcGIS, for example, to introduce the techniques of geo-vectorization in combination with your years of work experience.

　　First, the registration of the Basemap

There are different registration accuracy requirements for different scale maps:

1:200,000 geological map, the registration error can not be higher than 20 meters;

1:500,000 geological map, the registration error can not be higher than 50 meters.

Scanning quality of the original map, only need more than 4 control points with a single registration to achieve the accuracy requirements; The original map with bad scanning quality must be used two or three times, in ArcGIS, at least 6 control points can be two registration, at least 10 control points can be three registration. The control points are evenly distributed on the surface. For the convenience of registration, before registration, it is necessary to use image processing software for some of the serious aging of the map to remove the aging color, restore its primary colors.

　　Ii. establishment of a feature class (Feature Class) in a Geodatabase (geodatabase)

The establishment of the feature class is the main content of the establishment of the Geodatabase, in the vectorization work of geologic map, the main features are the polygon elements (Polygon Features), the line elements (lines Features), the point features (points fatures). Geology-related surface elements include: intrusive rocks, sedimentary rocks, metamorphic rocks, alteration zones, and various vein rocks; The main line elements are: faults, geological boundaries, etc. the main points are mainly various occurrences. Besides, there are non-geologic factors such as water system, roads, towns and settlements.

In the process of building a feature class, you want to set the name, alias, type (type), geometric feature (Geometry properties), coordinate system (coordinate system), XY tolerance (xy tolerance), Attribute table field (fields), and so on. Name is the basis of the computer Recognition feature class, its settings to be concise and clear, see the name of understanding, if necessary, you can also set the alias, please note that the alias is only for user identification, it is not the basis for computer recognition. Types (that is, polygons, lines, points), geometric features (that is, 3D data that does not include Z-values), coordinate system settings, and basemaps, to ensure that the vectorized data is consistent with the original data. XY tolerance should not be set too large, otherwise it will affect the accuracy of the data, we can be based on the specific accuracy requirements of the basemap set, such as 1:200,000 geological map, this value is set to 1 meters to meet the accuracy requirements. The setting of the attribute table field is important, so it is described in the next section.

　　Third, the attribute table of the feature class

The creation and filling of attribute table of feature class is the main and key content of vectorization work. The creation of a property sheet mainly includes the naming of fields and the selection of field types and lengths. For example, intrusive rocks can be used to set the local names of rocks, rock types, lithology symbols, lithology codes, rock colors, rock structures, rock formations, lithofacies, major minerals and content, secondary minerals and content, contact with surrounding rocks, contact surface direction, contact surface inclination, contact angle, surrounding rock age, formation era, ore-bearing and so on. For sedimentary rocks and strata, it is possible to set up the local name of strata, stratigraphic unit symbol, strata unit age, rock combination, the color of rock composition, main sedimentary structure of strata, biological fossils, production and so on.

field type, in the geological map vectorization commonly used mainly include: short integer, Long integer, Float, Double, Text, because of the characteristics of the field data selection.

field lengths are too short to be used, and are too long to cause data redundancy, especially for text types.

　　Vector tracking of points, lines and polygons

Point, line, surface vectorization tracking, must follow a basic principle, is faithful to the original image, consistent with the original.

Vectorization of points is easy, just place the points on the map to the appropriate size and position them at the center.

Line vectorization is also relatively simple, it is required to put the line to the appropriate thickness, followed by the mouse to track it. Since the scanned basemap will be a square grid when zoomed in to a certain extent, it is advocated to enlarge the basemap to a large multiple, tracking the mouse point in the center of the grid, so that the graph can be accurate to the grid.

In fact, such a statement is wrong, there are two drawbacks: one is slow, inefficient; second, the consequences of doing so may not be faithful to the original image, but will cause some minor errors. As shown in 1, the line of the basemap is originally smooth, and a lot of tiny jagged edges appear following a grid of one. It is important to realize that any raster image magnified to a certain extent will appear square raster, which is a big limitation of raster data, these squares are not true reflection of the original basemap, but the approximate expression of the original line. It is not the center of the grid that is to be traced, but the overall movement of all squares.

　　

Figure 1 Comparison of correct and incorrect tracking methods

A: The correct way to grasp the overall direction of all squares, smooth lines

B: The wrong way, followed by the center of each grid, there are a lot of tiny jagged

Thus, the so-called "meticulous" is relative, sometimes too meticulous, but will cause errors. So the correct tracking method should be: grasp the overall trend of all squares, so that the line to the maximum extent of this trend. In general, the lines should be smooth.

The vectorization of polygons is the most complex and cumbersome, and the workload is very large. In general, the topological relationship between polygons and polygons has the following four types: Absent, contiguous, intersecting, inclusive. To deal with the problem of vectorization is to properly handle the four topological relationships. The traditional method is: If two faces have a common side, in the construction of the use of capture, tracking and other tools to ensure that the two faces are strictly coincident on the public side. If a polygon contains another face, it is necessary to use a cutting tool when drawing, and then to create a new facet inside the large face.

It is clear that the above-mentioned Vectorization method in ArcGIS is a complete and general method. Such a method each encounter the common side to repeat the tracking, time-consuming and laborious, the operation is very cumbersome, a little careless will produce errors, the most common mistake is to produce some small triangles at the intersection of multiple faces. Such errors are also very cumbersome to modify.

It is precisely because of these shortcomings of the traditional method that we strongly recommend not to do this, but rather to use a more efficient method: The vectorization of all polygons as a line feature vectorization in line features, thus forming a line feature class called the original line of the drawing. Note: The bounds of all polygons must be closed when vectorization, and all lines can be tracked only once. Then use the "Feature to Polygon" tool in the toolbox to convert the original lines of the drawing into polygons. The advantage of this is that it is easy to operate, saves time and effort and greatly reduces the chance of error. 2 is shown below:

　　

Figure 2 Using the "Feature to Polygon" tool in the toolbox to convert any closed curve into a polygon

　　Five, The legend

This part of the operation belongs to the category of visual expression. ArcGIS has a powerful legend generation tool that can generate any desired legend, as long as you are familiar with the method of operation.

After the legend has been generated, it should be saved as a style file, which, once generated, can be modified and managed using style manger.

If you see a legend that you've done before, just call the style you used to do. You can also save the style in a complete set, and later encounter such a legend, just use the "Match to symbols in a style" method, so that you can save a lot of time. 3:

　　

Figure 3 Match to symbols in a style

　　Six, multi-person division of cooperation skills

If you want to complete a large scale in a short period of time, it is necessary to work together in a multi-person division. Multi-person collaboration requires that each member strictly enforce a uniform standard so that each part of everyone's responsibility is seamlessly integrated.

Before the start of the work, you can create a Division chart box, each of which is responsible for the part, when everyone's work is completed will be a map. As everyone is working according to the unified Division of labor chart frame, so the map when everyone's diagram can be well linked together. Some lines and polygons that are divided by the frame need to be restored to their original state using the merge tool. 4 is shown below:

　　

Figure 4 Division of labor collaboration, you can set up a Division diagram box, each of them complete one of them, when you use the Merge tool to eliminate the Division diagram frame Line

Division of cooperation requires a high degree of consistency, we have to implement a standard point to be refined to every aspect, such as vectorization tracking the distance between each node, when filling out the attribute table when the characters are uppercase or lowercase, is full-width or half-width, and so on. If you are unified in every detail, the chance of error will be reduced to very small, then the entire data will be seamless after the graph.

　　Vii. Meta-data

Some people call the metadata "data", which is equivalent to the data of the specification, for a complete set of data, metadata is essential. But the author of some vectorization staff survey found that the importance of metadata has not been enough attention, which is very inappropriate. For example, in shopping malls, without the product specifications we will think that its quality is not credible. Similarly, there is no metadata data, and its reliability is not too high.

ArcGIS provides us with a rich and powerful metadata editing tool, entering Arccatalog, selecting the data you want to edit, and then clicking the Data Editing tool to edit the metadata, and click on the tabs to edit the individual items. 5 is shown below:

　　

Figure 5 Meta-data editor

　　Summarize

Vectorization of geologic maps is a demanding and meticulous basic work, and high-quality vector data will provide reliable data support for future research. The author's Working group, before mastering the methods described above, each person with 6--7 days time to reluctantly complete a 1:200,000 of the regional geological map, master the scientific and efficient method, each person with 2--3 days can complete the same workload. This shows how important it is to master the scientific and efficient vectorization method.

ArcGIS Tutorial: Overview of vectorization Techniques for ArcGIS Geology maps