Image processing is among the complex and interesting areas of computer science, and its various developments have led us to impressive achievements and to improve our ability to convert images, correct and reduce interference or blurring, improve histograms of the image, change contrasts, reduce noise, magnifying or reducing images, compress images, and more.

Thanks to the information revolution, we are experiencing in the modern age, image processing is rapidly evolving and allows us to do much more than before in a variety of fields from the medical field to the security field.

What is deep learning, and how it has become the facilitator of a modern revolution that is changing our worldDeep learning is inspired by the information we have about the human brain functions. In a way similar to the brain, it uses a multilayered artificial neural network.Deep learning is considered to be one of the most important technological revolutions of our time, and takes a major part of  the AI ​​(Artificial Intelligence) revolution and the Information Revolution. These revolutions are born of the need to collect a very large amount of data that accumulates in the web, digital cameras, sensor data and through aggregate data storage. Its effects are evident in a variety of areas. The main ones:

• In the medical field – autonomic diagnosis, pathology identification, drug design, genetic research, risk prediction,
• In the field of quality of life – recommendation systems, virtual tour, robotics
• In the field of security and guarding – photography and surveillance, information gathering and intelligence analysis
• In banking – e-commerce, credit fraud exposure
• In the field of transport – streamlining public transport, autonomous driving
• In the field of entertainment – computer games, movies
• And more …

Deep learning and image processing
Advanced deep learning methods provide very precise and varied solutions to a variety of classical image processing problems. Some of them:

• Accurate sorting of multiple objects
• Identify the locations of the objects in the image
• Image segmentation
• 3D image analysis
• Creating images from scratch
• And more …

Apart from problem solving, the use of deep learning has given rise to advanced new functions, including the ability to artificially create images. For example, the use of generative adversarial network allows, among other things, to create images of faces, and the use of style transfer allows to change the style of an image in light of another image.