The basis of remote sensing digital image and the characteristics of remote sensing

1. the basis of remote sensing digital images

The most basic unit of remote sensing digital image is pixel, which is the sampling point in the process of remote sensing imaging. Each pixel has its spatial position characteristics and attribute characteristics, and the attribute characteristics are usually expressed by the brightness value (DN), which is determined by the electromagnetic radiation intensity detected by the remote sensing sensor. Figure 1 shows a remote sensing digital image. The brightness value of the original image is a dimensionless number, and the variation range is related to the quantization processing of the sensor. If the image quantization value is 8bit, the dynamic change range of the image brightness value is 0 ~ 255, and so on. According to the subdivision degree of sensors in electromagnetic spectrum, remote sensing images can be divided into single-band, multi-band and super-band images. Single-band images have only one brightness value per pixel, multi-band (also called multispectral) images have multiple brightness values, and super-band (also called hyperspectral) images have dozens or even hundreds of brightness values. Fig. 2 is a schematic diagram of a three-band digital image, in which the X direction reflects the number of columns in the image, the Y direction reflects the number of rows in the image, and the Z direction reflects the number of bands in the image.

The brightness value of remote sensing digital image is also related to the five resolutions of the image. These five resolutions are spectral resolution, spatial resolution, radiation resolution, time resolution and temperature resolution.

Spectral resolution refers to the ability and degree of imaging spectrometer to subdivide electromagnetic spectrum. The more bands, the higher the spectral resolution. For example, TM multispectral scanner has 6 bands with a bandwidth of 100~200 m, and imaging spectrometer AVIRIS has 224 bands with a bandwidth of 5~ 10nm. Generally, the more sensor bands, the narrower the band width, the more spectral information it contains, and the stronger the pertinence.

Spatial resolution refers to the size of the ground range corresponding to each pixel in the image. The smaller the range, the higher the spatial resolution. For example, the spatial resolution of TM multispectral scanner is 30m. The spatial resolution of panchromatic band of Quickbird image is 0.6 m.

Radiation resolution refers to the possible value or dynamic range of the brightness value of the recorded pixel. For example, the brightness value of an image pixel with a byte length of 8bit can be divided into 256 levels, and the brightness value of ground objects can only be one of 0~255.

? Time resolution refers to the frequency at which a sensor acquires images of a specific area. For example, Landsat takes 16 days to complete a global scan, while SPOT takes 3 days to complete a global scan. A small satellite group will visit any point on the earth in 12 hours.

Temperature resolution refers to the ability of thermal infrared sensor to distinguish the minimum difference of surface thermal radiation.

2. Characteristics of remote sensing

(1) Observing the Earth from space, and obtaining comprehensive surface information.

Remote sensing exploration obtains remote sensing data covering a large area in a certain period, comprehensively reflects the shape and distribution of various features on the earth, truly reflects the characteristics of geology, geomorphology, soil, vegetation, hydrology, artificial structure and so on, and comprehensively reveals the correlation between geographical things. In addition, because the detection band, imaging mode, imaging time and data recording of remote sensing can be designed as required, the data obtained by remote sensing are the same or similar. At the same time, considering that the new sensors and information records can be backward compatible, the data is comparable. Compared with the traditional ground investigation and investigation, remote sensing data can reflect the information of ground objects more objectively.

(2) Monitor the dynamic changes of ground objects in real time through the data of the detector platform.

Remote sensing detection can repeatedly observe the same area according to a certain period. For example, satellites in geosynchronous orbit can observe the earth once every half hour, and satellites in sun-synchronous orbit can observe the same area twice a day. This is helpful for people to discover and dynamically track the changes of many things on the earth through the obtained remote sensing data, and is conducive to studying the changing laws of nature, especially in monitoring weather conditions, natural disasters, environmental pollution, and even military targets. In contrast, the traditional ground survey needs to invest a lot of manpower and material resources, and it takes years or even decades to obtain the data of regional dynamic changes.

(3) Wide detection range, short data acquisition period and high speed.

Remote sensing detection can observe the earth in a large area from aviation or space platform in a short time, and obtain valuable remote sensing data from it. These data expand people's visual space and create extremely favorable conditions for grasping the present situation of things on the ground macroscopically. At the same time, it also provides valuable first-hand information for macro-study of natural phenomena and laws. Generally speaking, the higher the remote sensing platform, the wider the viewing angle, and the larger the ground area that can be detected synchronously, the easier it is to find some large and important targets on the earth and their spatial distribution laws.

(4) Massive information can be obtained through various channels, with many benefits.

Compared with traditional methods, remote sensing can greatly save manpower, material resources, financial resources and time, and has high economic and social benefits. It is estimated that the ratio of economic investment to benefit of American Landsat is 1: 80, or even higher.

It is not difficult to see that remote sensing technology has the advantages of comprehensiveness, short cycle, real-time, wide coverage, macro and good benefit. For the application of remote sensing, with the continuous improvement of satellite image resolution (space, time and spectrum) and the innovation and perfection of image processing technologies such as image correction, enhancement and fusion, satellite images are applied to weather forecast, marine monitoring, environmental monitoring, geological survey, land survey, forest farmers survey, resource management, urban planning, hydrological observation, topographic mapping, disaster monitoring and evaluation, development and planning of scenic spots, and Remote sensing serves the national economy and social development. Remote sensing of the earth is one of the main application fields with the earth's resources, environment and disasters as the main research objects. Environmental remote sensing can monitor, evaluate and predict the status and dynamic changes of nature and environment. Due to the growth of population and the development and utilization of resources, nature and environment are changing at any time. Using the characteristics of multi-temporal and short period of remote sensing, it can quickly provide reliable basis for environmental monitoring, evaluation and prediction. The application research of remote sensing can be divided into global remote sensing, regional remote sensing and urban remote sensing on the spatial scale.