Computer vision libs and techniques for Ramudroid

Raspberry Pi camera module

Raspberry Pi Camera find applications in areas from robotics to home automation. applications provided: raspistill, raspivid, raspiyuv and raspividyuv

Deep Learning

Google’s TensorFlow is a python library for Deep Learning. It can be used for image recognition, face detection, natural language processing etc 2 methods to install TensorFlow on Raspberry Pi:

• TensorFlow for CPU
• TensorFlow for Edge TPU Co-Processor

Install deep-learning libraries tensorflow and keras

pip3 install tensorflow
pip3 install keras

OpenCV

Opencv (cv2 lib in python ) is used for computer vision , it captures and trabsforms the pictures. numpy and matplotlib are use for image pre-processing and rendering along with it

sudo apt-get install libhdf5-dev -y && sudo apt-get install libhdf5-serial-dev -y && sudo apt-get install libatlas-base-dev -y && sudo apt-get install libjasper-dev -y && sudo apt-get install libqtgui4 -y && sudo apt-get install libqt4-test -y
pip3 install opencv-python
pip3 install matplotlib

Models

Retina net

wget https://github.com/OlafenwaMoses/ImageAI/releases/download/1.0/resnet50_coco_best_v2.0.1.h5

Run

To detect an object and draw box to identify it

python3 detect.py

sample results in outputs folder

3x3 grid , image detection on centre box

dividing image into 3x3 grid

sample

perform image detection on box-5

3x2 grid , edge detection on bottom centre box

dividing image into 3x2 grid

sample

Canny Edge Detection

Perform edge detection on box-5

Initial edge detection using canny edge method doesnt give satisfactory result in evaluating edges

Edge Detection after adding sharpness filter using GaussianBlur

Ref

• https://circuitdigest.com/
• https://github.com/opencv/opencv
• https://docs.opencv.org/master/d9/df8/tutorial_root.html
• https://learning.oreilly.com/library/view/computer-vision-with/