**3.6 Geochemical image classification**

In the vector image, undoubtedly, the anomaly image of elements is one of the final products in geochemistry. The anomaly map of elements may give users vivid

**Figure 8.** *K2O, Na2O, and SiO2 ternary color image synthesis in the middle segment of Daxinganling metallogenic belt.*

**61**

**Figure 10.**

*(yellow line), and 95% (red line).*

*The Geochemical Data Imaging and Application in Geoscience: Taking the Northern…*

*Numerical statistical histogram of Na2O content in the middle segment of the Daxinganling metallogenic belt.*

visual impression. Thus prospecting researchers can directly use the geochemical anomaly maps to explore the interested target. The results expressed in the rasterized image can also be fully used so as to employ the statistical method of density slicing. **Figure 10** is a density sliced map which was created by histogram statistics of copper element, and its result is similar to the geochemical anomaly map. Their difference is that the final rasterized image was irregularly dentate if enlarging a small area. What is mentioned above is the simplest classification in the rasterized geochemical image, and they were based on the sole element anomaly. Most of the time, the classification using remote sensing images is divided by supervised one and unsupervised one, and their difference is that the supervised classification

*Anomaly map of copper element formed in density slice in the middle of the Daxinganling metallogenic belt. According to histogram cumulative frequency statistics, anomalies were graded to 75% (green line), 85%* 

*DOI: http://dx.doi.org/10.5772/intechopen.84725*

**Figure 9.**

*The Geochemical Data Imaging and Application in Geoscience: Taking the Northern… DOI: http://dx.doi.org/10.5772/intechopen.84725*

**Figure 9.** *Numerical statistical histogram of Na2O content in the middle segment of the Daxinganling metallogenic belt.*

#### **Figure 10.**

*Applied Geochemistry with Case Studies on Geological Formations, Exploration Techniques…*

sification based on histogram analysis will be introduced in the next step. Density slices to a gray geochemical image can create element anomalies. Cumulative frequency percentage can be used to determine anomalies or anomalies

mean value, median, mode, range, contrast, etc.

**3.5 Algebraic operations and logic operations of image**

grading (**Figure 10**).

maps can be directly performed.

**3.6 Geochemical image classification**

tion could be generated.

all the pixels within the image. Basic statistics of a geochemical image involves the

Histogram is one of the important statistics of a geochemical image. Histogram refers to a discrete graph of probability density function of all gray values in the image, or it may be seen as a graphic expression of basic statistics of gray image. **Figure 9** is based on histogram and the chiefly related statistics. Under ENVI software, the calculation results of cumulative frequency can be obtained, and clas-

Algebraic operations of image indicate that the corresponding image pixels of two (or more than two) of input images received four arithmetic operations, which in order are addition, subtraction, multiplication, and division. The algebraic operation cannot be directly fulfilled within the vector maps, while the rasterized

Logical operations of images are widely applied, for instance, the masking method mentioned above used logic operations to form a mask band. A specific value in a pixel could be obtained by logical operations, and then a simple classifica-

In the vector image, undoubtedly, the anomaly image of elements is one of the final products in geochemistry. The anomaly map of elements may give users vivid

*K2O, Na2O, and SiO2 ternary color image synthesis in the middle segment of Daxinganling metallogenic belt.*

**60**

**Figure 8.**

*Anomaly map of copper element formed in density slice in the middle of the Daxinganling metallogenic belt. According to histogram cumulative frequency statistics, anomalies were graded to 75% (green line), 85% (yellow line), and 95% (red line).*

visual impression. Thus prospecting researchers can directly use the geochemical anomaly maps to explore the interested target. The results expressed in the rasterized image can also be fully used so as to employ the statistical method of density slicing. **Figure 10** is a density sliced map which was created by histogram statistics of copper element, and its result is similar to the geochemical anomaly map. Their difference is that the final rasterized image was irregularly dentate if enlarging a small area.

What is mentioned above is the simplest classification in the rasterized geochemical image, and they were based on the sole element anomaly. Most of the time, the classification using remote sensing images is divided by supervised one and unsupervised one, and their difference is that the supervised classification

firstly gives category, whereas the unsupervised one is determined by the statistics characteristics of image data itself. The classification method used for remote images are suitable for the geochemical atlas. Usually employed methods include multilevel slice classifier, decision tree classifier, minimum distance classifier, maximum likelihood classifier, and the like (e.g., method of fuzzy theory, expert system method, etc.). SAM method mentioned later is one of the supervised classification methods.
