Processing Image:

           Processing any kind of function using coding is simple as explaining or instructing a normal human to do task, the difference is here that here is that programing language is only way to communicate. In Image processing the minute detail are needing to make the image displayed correctly to the screen. The first day we went over some old stuff, of how to utilize the git lab, setting up virtual environment for python. The library we are using is OpenCV, Matplotlib, numpy.  The process starts when user uploads an image, which uses the OpenCV function ‘imread’ to read the image, there are three ways you can read an image Either in Gray scale, Color or luma. Since the image is bigger the size from the screen one need to fit the image according to the size of the dimensions of computer screen. Make sure one gets the correct value to resize, or use calculator if you get wrong values can get you something like following image size and result will also be not even close to accurate:

. 

Once you get the correct dimensions of your computer screen the image size will be full screen and object detect will be 100% accurate. Something like following:

 After that we need to clear the image thought the process called ‘smoothing’, where we separate the noise from the image, Now since we are not trying to separate color information in the image, we do not need ‘dilation’ of the image, means we do not need to specify the RGB pattern we look in the image. Instead we process the image more by match the sample image/object which is a standard image as an reference to the object we look in the image.  Yet the process we did so far is not detecting any object since we never programmed the algorithm to detect a object in the image. The above process helps us to understand the difference in image processing and Object recognizing. 

What do we need to know to build the Image Recognition Library:

                  

            Building a library is a same concept of writing books, A book has a lot of information and knowledge. Which can be used in developing some great ideas, later using the knowledge one got from reading a book. If We combine all the related topic or field books it will become a library for that topic or field, same is with building image recognition library, multiple or single person develop an algorithm and uses it and code a standard form of program that works as multipurpose ways

         In most of the time one can code simple function of AI without knowing anything about it, in that case the program will be just that function. Or One can also utilize the existing Basic AI library and use the pre-built methods and functions, Such method will give you close result but mostly not close to the accuracy that one wants.

         In final words, one may or may not need to know programming that's a different talk, but most what one need is to understand what and how system needs to function to the task given to it.