👨🏾‍💻- AI-Integrated NFT - Technical Analyse

1. Colour (the three primary colours)

Colour is a visual effect to light generated from our eyes, brain, and life experience. Human eyes are most sensitive to red, green and blue. And the majority of colours can be created when the three primary colours are added together in different combination and quantities. In the same way, most of the monochromatic light can be separated into red, green and blue colour lights. This is the most fundamental principle of colorimetry, also known as the “three primary colour theory”. Using a glass prism, Issac Newton became the first person to discover that natural light can be split into 7 different colour lights. Newton believed that since the natural light could be split, then the 7 colour lights should also be able to be further separated or combined. Among the 7 colour lights, red, green and blue are the only three colours that can not be further split into other colours. This is why they are called the “three primary colours”, and the rest of colour lights can be calculated through an additive and subtractive mixture of the three colour lights.

All the colours we have seen in nature are derived from the combination of the three primary colours. Each of the primary colours will have a value to represent its intensity. Computers today store colour value in 8-bit, and the range of each colour value is 0-255. Based on the principle of additive mixture, there will be 255,255,255 kinds of colour when the three primary colours are combined, which is sufficient for the resolution capability of human eyes. In RGB colour model, colours are represented by three values that respectively indicate their intensity. The minimal intensity value is 0 and the maximum value is 255. When all three colours are 0 , then the black colour is shown, and the white colour is generated when the colours are all 255. We might as well check the RGB colour codes charts (a different RGB value corresponds to a different code).

2. Pixel

The pixel is defined as a small square graphics in a colour that makes up a digital image. All of these small squares have a clear position and a colour value allocated to each of them. And it is the colour and position of this pixel that determine what an image will look like. Pixel can be deemed as a non separable unit or element of a graphics, which means a pixel can not be further divided into a smaller unit or element. A pixel exists in a single colour square and each of a bitmap graphics encompasses a certain amount of pixels, which determine the size of an image on a display.

3. Image Computation

By enlarging an image on the screen through a zoom, it is possible to observe the countless pixels that allow the creation of the picture. In fact, everything originates from an infinite scale in the first place and ends up a limited entity afterward. For instance, a plane is made of countless dots, and when the pixel pitch becomes tight enough, it would be impossible for us to distinguish the individual pixels with our naked eyes-all we can see is an image. Another example is that if there are countless images flashing in front of our eyes, we will be watching a movie. They all share the identical principle just like in geometry-a line is a collection of infinite points, a plane consists of infinite lines, and a three-dimensional space is made of planes. If we are able to appreciate what has been analysed, it would be easy enough to deal with image recognition: by simply programming a computer to calculate the characteristics of the countless dots that form the picture, all images could be recognised. This, however, requires powerful computing capacity as the high resolution images consist of huge numbers of pixels. Take a laptop for example, its screen usually has a resolution of 1366*768 dots, which means there are millions of pixels on it. And if it is in RGB model, the colour values will be even higher than usual.

4. Deep Learning

Deep Learning (DL) is a neural network designed to imitate and analyse the workings of the human brain. It is also a machine learning technology that simulates the operation pattern of human brain in processing and interpreting data. The fundamental feature of DL is the pattern of imitating the nerve cell transmission and information processing function of human brain. DL is widely known for its application in computer vision and natural language processing (NLP). It is evident that DL is closely related to the “neural network” of machine learning. And the “neural network” is the main algorithm and method of DL; or it could be named “Deep Learning” an algorithm of “improved version of neural network”.

The “Da Vinci Algorithm” will be based on the pixels and the colour value of images as well as the combination of human emotions expressed in his works, which enable the algorithm to conduct in-depth analysis and multidimensional learning to art works, and build a complex and anthropomorphic neural network.

5. NFT encryption art generation

Based on the learning model that has been established and the creation needs(emotions) proposed by human, the “Da Vinci Algorithm” is capable of generating the NFT encryption art that fulfils the needs through anthropomorphic thinkings and analytical judgement as well as the complex computation of hash rate.