A (Clear) Line Between Computer Vision and Image Processing

I have been working in and learning about both Computer vision and Image processing for a few years now, and I believe that I am not a complete beginner any more.

Still, after all these years, it is hard for me to tell for any particular part of my work whether it is mostly Computer vision related, or if it is Image processing. I just can't see the line -- when I work, study and research, I read reference materials with both keywords.

So, I am interested in the definition of the Computer vision and Image processing fields, with focus on the comparison (differences as well as overlaps) of the fields.

Additionally, I think it would be valuable to have an example of a (conceptual or existing) practical applications, projects and works dealing with/utilizing:

• solely (or mostly) Computer vision tools and ideas


• solely (or mostly) Image processing tools and ideas


• a combination of tools and ideas from both fields

with special attention paid to what makes it one and not the other, or what makes it both.

I understand that these fields are highly related and that the "line" might not be as clear as this question is asking for, but I hope you understand that the point of this question was not to design a simple decision rule for classifying (my) work, but rather a better understanding of the focus and goals of these fields. Also, any additional information that seems on-topic with my question tone is welcome, even if I did not specifically ask for it.

Answer

I believe Gonzalez and Woods are competent enough to rely on their opinion:

There is no general agreement among authors regarding where image processing stops and other related areas, such as image analysis and computer vision, start. Sometimes a distinction is made by defining image processing as a discipline in which both the input and output of a process are images.We believe this to be a limiting and somewhat artificial boundary. For example, under this definition, even the trivial task of computing the average intensity of an image (which yields a single number) would not be considered an image processing operation. On the other hand, there are fields such as computer vision whose ultimate goal is to use computers to emulate human vision, including learning and being able to make inferences and take actions based on visual inputs. This area itself is a branch of artificial intelligence (AI) whose objective is to emulate human intelligence. The field of AI is in its earliest stages of infancy in terms of development, with progress having been much slower than originally anticipated. The area of image analysis (also called image understanding) is in between image processing and computer vision.

So I would say, that primary difference is in goals, not methods. For example, if the goal is to enhance image for later use by humans, than this may be called image processing. And if the goal is to emulate human vision (be it object recognition, defect detection or automatic driving), then it is closer to computer vision. Note, however, that emulating human vision by definition may also require image enhancement, so in most real cases computer vision relies on image processing.

Image understanding (feature extraction) may be equally used in both - pure image processing and computer vision.