Basic attributes of remote sensing images

The spectral characteristics, spatial characteristics and temporal characteristics of remote sensing images are important research contents of remote sensing geology.

Spectral characteristics

From the perspective of spectroscopy, it is the basic starting point of remote sensing application to identify ground objects and phenomena according to the difference of spectral characteristics recorded by remote sensors. The difference of spectral characteristics in remote sensing images is the difference of image gray (tone) or color. The gray scale or color of various remote sensing images is the reflection of electromagnetic radiation energy in its response band: black-and-white panchromatic photos and natural color photos reflect the reflected energy of ground objects to visible light (0.38-0.76μm); Black-and-white infrared photos and color infrared photos reflect the reflected energy of ground objects in some visible light and photographic infrared bands (0.38- 1.3μm). Thermal infrared image reflects the thermal radiation energy (radiation temperature) of ground objects in thermal infrared band (8- 14μm); The imaging radar image reflects the intensity of backscattering echo of ground objects to artificially emitted microwave (0.8- 100cm); The gray scale of multi-band and ultra-multi-band images is the reflection of the radiation energy of their respective response bands. Spectral characteristics analysis of remote sensing images, including spectral resolution and radiation resolution of remote sensors.

(2) Spatial features (geometric features)

The spatial characteristics of remote sensing images are an important basis for morphological recognition, mapping, establishment of interpretation marks, geometric correction and enhancement of images. The analysis of spatial characteristics of remote sensing images mainly includes spatial resolution, image projection attributes, scale, geometric distortion and so on.

1. Spatial resolution

Spatial resolution of remote sensing images refers to the ability of images to distinguish adjacent targets with different contrast and a certain distance apart.

(1) Image resolution: refers to the logarithm of black and white lines (line pair /mm) that can be resolved within the width of 1mm when observing an image with a microscope. It is influenced by the resolution of optical system, the resolution of photosensitive material (or display), the image scale and the contrast between adjacent objects.

(2) Ground resolution: refers to the minimum distance between two objects that can be resolved in remote sensing images. The scanned image is usually represented by the temporary field of view of the remote sensor detection unit. For example, the MSS image of Landsat has a ground resolution of 80 meters.

2. Image scale

Refers to the ratio of the length of a line segment on the image to the corresponding horizontal distance on the ground. It is determined by the ratio of the focal length (f) of the optical system of the remote sensor to the height (h) of the remote sensing platform, that is, 1/m=f/H (Figure 3-27). Because remote sensing images are generally central projection or multi-center projection, which is different from the vertical projection of maps. The image scale is affected by the terrain fluctuation and the position of ground objects in the image frame, and there will be inconsistencies everywhere.

3. Projection characteristics and image geometric distortion

Remote sensing images are all focused by optical system, and the projection properties of remote sensing images are determined by the imaging law of lens and the imaging mode of remote sensor. Different projection attributes will produce different geometric distortions of images.

(1) Central projection: As shown in Figure 32-7, the light rays (Aa, Bb, Cc) projected by various objects on the ground pass through a fixed point (S) and are projected onto the projection plane (P 1, P2), and the formed perspective image is called central projection, and S is called projection center (lens center). Frame photography is the central projection of the ground. When the projection center is located between the projection plane and the ground object, the perspective image (P 1) on the projection plane is called negative, and P 1 is called negative. The image on the projection plane (P2) between the projection center and the ground object is called positive image, and P2 is also called positive film (photo). The intersection of the main optical axis of aerial camera and the image plane is called the image principal point; The intersection of the vertical line passing through the projection center and the image plane is called the image bottom point.

(2) The one-dimensional central projection banner camera image is the central projection along the gap direction. When the ground is flat and the projection plane is horizontal, the image scale is equal to f/H, but in the heading direction, the proportional relationship is determined by the ratio of film winding speed v to speed v, so the vertical and horizontal scales of the image are usually inconsistent. Panoramic photographic images also belong to one-dimensional central projection when the scanning angle is active, which will produce panoramic distortion (see panoramic photo characteristics).

(3) Multicenter projection: The images scanned by mechatronics are scanned point by point, and each image point has its own projection center. However, the imaging time of each image point on the same scanning line is very short, so it can be considered that each scanning line has a projection center, so the image scanned by mechatronics (stereo) is a multi-center projection (Figure 3- 16).

(4) Rotating oblique projection: as shown in Figure 3-28, it shows the geometric relationship when the side-looking radar images the flat ground, Sab is the image plane, ab is the trajectory of the light spot on the screen of the cathode ray tube, and the time when the light spot appears depends on the time interval between the radar transmitting microwave and receiving echo. Because the microwave propagation velocity (c) is fixed, the radar image is actually oblique projection, and the projection property is rotational oblique projection.

Figure 3-27 Central Projection

Figure 3-28 Rotation Oblique Projection

(3) Time characteristics

Remote sensing image is a record of electromagnetic radiation energy of ground objects at the moment of imaging, and all ground objects change with time. One is the process of natural change, that is, its occurrence, development and evolution; The second is rhythm, that is, the development of things shows some periodic and repetitive laws in time series, that is, the spectral characteristics of ground objects change with time. In the interpretation of remote sensing images, we must consider the time and position of the research object, make full use of multi-temporal images, and can't cover the whole development process with one instant information. The temporal characteristics of remote sensing images are related to the temporal resolution of remote sensors, as well as the imaging season and time.