Introduction to each band combination of TM imagery

321:
True Color synthesis, that is, 3, 2, 1 bands are given red, green, blue, then the natural color synthesis image, the color of the image and the original region or the actual color of the scene, suitable for non-remote sensing application professionals to use.

432:
Standard false color synthesis, that is, 4, 3, 2 bands are given red, green, blue, to obtain the image of vegetation into red, due to the prominent performance of the characteristics of vegetation, the application is very broad, and is called standard false color.
Example: Satellite remote sensing image shows the outbreak of cyanobacteria
Let's take a look at the mechanism of remote sensing monitoring in cyanobacteria outbreak. The green algae organism mixed with the white foam pollutants gathered in the water surface, the spectral characteristics of cyanobacteria coverage were obviously different from those of the surrounding lake. Due to the role of high chlorophyll a, the cyanobacteria region has a high reflectivity in the LandsatTM2 band, and the reflectivity of TM3 band is slightly lower but still higher than that of the lake, and the reflectivity of the TM4 band reaches the maximum. Therefore, in TM4 (red), 3 (green), 2 (blue) false color synthetic images, the cyanobacteria area is crimson, and the surrounding dark blue, blue and Black Lake is clearly different. In addition, the algal blooms are affected by the flow and wind direction of the lake, showing a stripe extension, a striped structure and a floc texture on the TM image, and a distinct difference from the surrounding lake surface.

451:
The most informative combination, TM image of the light wave information has a structure of a-dimensional, the physical meaning of the equivalent of brightness, green, heat and humidity. In the spectral images of TM7 bands, the general 5th band contains the most abundant information of the figures. Between the 3 visible bands (i.e., the 1th, 2, 3 bands), the high correlation between the two mid-infrared bands (i.e., the 4th and 7 bands) indicates considerable repeatability or redundancy in the information of these bands. The 4th and 6 bands are more special, especially the 4th band has very low correlation with other bands, which indicates that the band information has great independence. The results of calculating the entropy of various combinations show that the color synthetic images composed of a visible band, a mid-infrared band and a 4th band are generally the most abundant information of the figures, in which the combination of 4,5,3 or 4,5,1 bands is often the best. The 7th band is only in the detection of forest fire, rock and mineral alteration zone and soil clay mineral types and other aspects have a special role. When the optimal band combination is selected, it is necessary to consider the color assignment problem to obtain the best color synthetic image. The most sensitive color of the human eye is green, followed by red, blue. Therefore, green should be assigned to the band with the greatest variance. According to this principle, the use of 4, 5, 3 bands, respectively, Red, green, blue composite images, color contrast is obvious, rich levels, and the color display of various types of objects similar to conventional composite films, in line with the past conventional film visual interpretation habits. For example, the 4, 52-band color Swap, that is, 5, 4, 3 are given red, green, and blue, the approximate natural color synthesis image, suitable for non-remote sensing application professionals.

741:
Band combination image has the advantages of compatible with mid-infrared, near-infrared and visible-band information, with rich color, good layering, rich geological information and surface environment information, high definition, less interference information, high degree of geological interpretation, various structural traces (folds and fractures) are shown clearly, The boundaries of different types of rocks, the boundaries of rock formations, the distribution of special lithology, and the volcanic structures are also clearly shown.

742:
In 1992, the comprehensive interpretation of remote sensing geology of gold and silver mineralization in southeastern Guangxi was completed, using 1:100,000 TM7, 4, 2 false color composite to interpret the linear structure 1615, the annular image 481, and summarized the tectonic alteration rock type, quartz vein type, On the basis of the remote sensing image characteristics and metallogenic model of volcanic type typical deposit, the metallogenic prediction of the whole area into the hall is delineated, the gold and silver a metallogenic prospect area 2, Class B 4, and Class C 5. It provides remote sensing basis for optimizing prospecting target area in this area.

743:
China has successfully monitored the forest fire and post-disaster changes in Daxing ' anling by using the United States Landsat thematic mapping instrument image. This is because the TM7 band (2.08-2.35 microns) is sensitive to temperature changes, the TM4, TM3 bands belong to the infrared and red regions respectively, can reflect the best band of vegetation, and has the function of reducing the smoke effect, while TM7, TM4, TM3 (respectively, given red, green, Blue) color synthesis of colors close to the natural color, it can be through the analysis of TM743 color synthesis map to command the spread and control of forest fires and the recovery of the forest after the disaster.
754: For different periods of lake water level changes, can also be used in different bands, such as the use of terrestrial satellite MSS7,MSS5,MSS4 synthetic standard false color images of blue, dark blue and other different levels of color can be distinguished. Thus, it can be used to analyze the geographical regularity of lake water level change.

541:
The remote Sensing survey of the sandstone ore deposit in XX Development Zone is based on the analysis of the selection of Fefee7 (TM5, TM4, TM1) and various remote sensing data of aviation and spaceflight in the Landsat TM Best Band group, and the fourth system landform in the investigation area is preliminarily interpreted. For example, the 4, 52-band color Swap, that is, 5, 4, 3 are given red, green, and blue, the approximate natural color synthesis image, suitable for non-remote sensing application professionals.

543:
band selection and principal component analysis our research uses the TM data of August 2, 1995. For screen display and screen image analysis, choose the most informative 5, 4, 3 band combination with red, green, blue three colors to generate false color composite images, the composite image is not only similar to the natural color, more symbolic of people's visual habits, and because of the rich information, can fully display the different features of the image of various objects , so as to facilitate the selection of training venues, can ensure the accuracy of the training site, for computer Automatic identification classification, the use of principal component analysis (K-L transform) data compression, the formation of three components of the image data for automatic identification classification. This work is a comprehensive method, which is based on the interpretation of remote sensing image and the combination of geological and geophysical exploration data. Interpretation for visual interpretation, the interpretation of remote sensing images are: The March 1984 imaging Processing amplification to 1:50,000 satellite TM pseudo-color film (5, 4, 3-band synthesis) and July 1979 shot 1:16,000 black and white aerial film as the main work of the film; using November 1986 1:100,000 TM Fake color film (7, 4, 2-band synthesis) for the reference film species.

453:
The remote sensing information source is the Chinese Academy of Sciences satellite remote sensing ground receiving station in October 1995 received the United States MSS Satellite Remote Sensing TM Band 4 (red), Band 5 (green), Band 3 (blue) CCT tape data produced 1:100,000 and 1:50,000 false color synthetic satellite imagery. The landscape of karst landform, such as mountainous area, hilly area and plain platform, and different types of land-use images are distinct. During the imaging period late rice is close to harvest, and there is no stagnant water in the paddy field, so the paddy fields in the cultivated land are pink, the dry land is mostly harvested, and the soil moisture is low, and it is gray and white; the vegetable field is bright and scarlet because the vegetables grow well. The garden tones are light brown and the plots are orderly and clear in shape. The arbor tones in the woodland are dark brown, while the shrubs distributed in the karst hilly areas are yellow to tan. Most of the pasture is in yellow and green tones. The towns in the construction land are blue, the highways are line-like, the lines are gray, the line is blue, the airport runway is bluish straight, the background grass is blue-green, the construction of the new airport is white and the old airport is white with a clear outline and a more regular shape. The reservoirs and rivers are in dark blue tones. Take 4, 5, 3 bands, respectively, Red, green, blue composite images, color contrast is obvious, rich levels, and the color display of various types of objects similar to conventional synthetic tablets, in line with the past conventional film visual interpretation habits. 472: In the use of TM4, 7, 2-band false color synthesis and 1:4 computer interpolation amplification technology, in the production of 1:50,000 TM image map and 1:5 million engineering geological map, the development rate of the bank collapse of the quantitative monitoring and on a single sheet of air on the Measurement of Rock (broken) layer production, etc., are unique.

Type extraction:
1. Urban and township extraction: TM1+TM7+TM3+TM5+TM6+TM2-TM4
2. Township and village: TM1+TM2+TM3+TM6+TM7-TM4-TM5
3. Extraction of rivers: TM5+TM6+TM7-TM1-TM2-TM4
4. Road extraction: tm6-(TM1+TM2+TM3+TM4+TM5+TM7)

Spectral differences
TM1 residents and river vegetables are not easy to separate.
TM2 residents and river plots are not easy to divide
TM3 countryside and vegetable garden are not easy to divide
TM4 farmland and road is not easy to divide, township, road, floodplain easy muddy.
TM5 County and farmland are not easy to divide
TM6 Village and River easy to mix

The difference between false color images and other images
Usually the single-band or panchromatic bands in RS are black and white, and the quality of black-and-white images is generally measured in gray scale. The three-band combination is represented as a color image, including:
True Color (True Color): (three-band combination), respectively, the RGB three-band image is given RGB three colors, one by one, the color of the composite image and the original region or the actual color of the scene, called True Color, True color is the only synthesis.
Pseudo-color (pseudo color): The black and white image is transformed into a color image, different shades of gray or gray range by the value of different colors or a series of colors, the color of the image and the actual color is inconsistent, that is, pseudo-color image.
False color (false color): (three-band combination), to the different bands of the image is given RGB tri-color, not with the original band RGB three bands one by one corresponding to the image color and actual color is inconsistent, called false color image, false color image is to make some features more obvious, Helps us to interpret and analyze.

The different combinations of TM any three bands (except over 321 combinations) Form different pseudo-color representations and different applications:
3,2,1
This RGB combination simulates a pair of natural-color images. Normal color image. Suitable for shallow sea exploration and mapping. Sometimes used for shoreline research and smoke plume detection.
4,5,3
for analysis of soil moisture and vegetation conditions. Also good for the determination of inland water bodies and land/water boundaries.
4,3,2
infrared false color, infrared color image. Provides a medium spatial resolution. In this combination, all vegetation is shown in red. This is the most commonly used band combination for remote sensing of vegetation, crops, land use and wetland analysis.
7,4,2
suitable for temperate to arid areas. Provides maximum spectral diversity. Soil and vegetation moisture content analysis, inland water body positioning. The vegetation is shaded in green.
5,4,3
distinction between urban and rural land use, determination of land/water boundary.
4,5,7
detects clouds, snow and ice (especially in high-dimensional areas).
4-3/4+3 ndvi-standard deviation vegetation index; different ratios of TM band 4:3 have been shown to be useful in enhancing the contrast of different vegetation types.
7,5,4
http://bbs.esrichina-bj.cn/esri/viewthread.php?tid=125066

Introduction to each band combination of TM imagery