Three levels of digital image processing

Digital image processing is divided into three levels: low-level image processing, image processing intermediate and advanced image processing (narrow image processing, image analysis and image understanding).

 

Narrow image processing: a transformation of the input image to obtain an output image, an image is to process the image.

(1) lower the content of image processing (image processing narrow)

SUMMARY: Various image processing mainly to improve the visual effect of the image or projection of useful information, and laying the foundation for the automatic recognition, or by encoding to reduce the required storage space requirements thereof, transmission bandwidth or transmission time.
Features: transformation is performed between the input image, the image is output, i.e., the image.


(2) Intermediate image processing (image analysis)
Content: The main interest of the target image is detected (or split) and measured, to obtain objective information thereof, so as to establish a description of the object in the image, from the image to a Numerical or symbols. .
Features: the input is an image, the data is output.

(3) Advanced Image Processing (Image Understanding)
Content: Based on the mid-level image processing, further study the nature of each object in the image and mutual connections between them, and come to understand the meaning of the content of the image (object recognition) and the original objective interpretation of the scene (computer vision) to guide planning and action.

Features: The objective world as the center, with the knowledge and experience to grasp the whole objective world. "Is the data input, the output is appreciated

 

Differences and connections: narrow image processing is a low-level operations, it is mainly processed on the image pixel level, an amount of processed data is very large; image analysis proceeds to the middle, extracting segmented and wherein, the original image of a pixel constituting converted into relatively simple, non-image forms of description; high level operation of an image understanding, it is described in the abstract reasoning symbols, and its processing method reasoning human thinking many similarities.

Guess you like

Origin www.cnblogs.com/h694879357/p/12551723.html