implimentation real time figer count system using openCv an ML project
BY ENMBE VERINE ICHUMBEI 06-10-2021
INTRUDUCTION The year was an exciting one as i had mainly two objectives as model 1 of the techcam offered by inchtec . AI and Data Analysis both interesting sections the program was scheduled for two weeks in which were to be in attendance twice each week Tuesdays and Thursdays 8am to 12am. The program was distributed among three instructors. Data analysis. bellow we shall look at content of work
CONTENT OF WORK Our content was not so tight but essential cause i would say the content gave me a lot of grounds to be able to take on bigger projects. to the best of my knowledge our instructors try to make us see in each of our lesson the relationship between the working concept and it real life applications in solving problems. After a briefer introduction of the history of Ai and data analysis we went on to the work proper
AI • Opencv An intoduction to opencv as one of best libraries fo vision(detection, recondition and tracking) only draw backs is it use cascade which is found of false negative(picking up unneeded background object ) but never the less it quit fast
• Image reading with opencv and python Our case study was phase detection. Here we were to see we could use opencv to get and imge from the computer (read) we could also do the same for a video in the computer using the VideoCapture("Video.mp4") face detection on an image : face detection on a video:
• Real life image capture with opencv and python here we made use of accessing an image through our webcam to identify a phase by analyzing the different farms so the phase detection was don using a webcam Face detection real time:
• IPwebcam with opencv and python The challenge here was to capture an image detect a phace and save the image maybe this was done by routing the signal though a link pick up by the VideoCapture("link") also with ip webcam in the mobile phone device
Project This is practically my best part design a program to detect a hand, detect the number of fingers up, and display on the screen the number to do so . Well the project was cool
Challenges: my main goal was how i would tell the computer that when my one finger was up it means one so i made some research and found out opencv had a function and an assistant library for that "mediapipe" i mean that was actual the solution.
Plane: • Create a module called handcrack.py it had thre main methods
- A constructor to handle all the constants
- HandFinder to detect the presence of a hand perform processing to identify the fingers
- Findcoord.To draw up a hand done by processing we need land mark but only the processing function had that but i need to tell if a finger is up by comparing it the other so that is why we needed to know the coordinate position of each land mark at any time this coordinate are to be stored in a list for later use
• Create the main file called finger_count.py to handell all operation of our program with the help of our handcrack.py imported.
Conclusions In all i would say we accomplish a grate task and dou they fully meet my expectation the sure did offer me a lotto that i say thank you