Human body measurement

Human body measurement is a image processing software use in identify body parts, calculate body measurements.

Take your picture in one color background
Enter your height and choose input image path
Magic!!!

Version

0.0.1

Tech

Human body measurement uses a number of open source projects to work properly:

NumPy - The fundamental package for scientific computing with Python
Scikit-image - Image processing in Python

Test and view tools

IPython
IPython notebook

Installation

Python: sudo apt-get install python-pip python-dev build-essential
Dependencies: pandas, sympy, nose. (pip install 'lib-name')
Numpy: sudo pip install numpy
Matplotlib: sudo pip install matplotlib
Scipy: sudo pip install scipy
Tool (ipython, ipython notebook): pip install ipython, pip install "ipython[notebook]"
Scikit-image: pip install -U scikit-image
How to use ipython-notebook: https://www.youtube.com/watch?v=H6dLGQw9yFQ

Procedure For Image Capture

Lighting: Bright environment with as much natural light as possible
Background: Removal of obstructions with a single color background
Border: Encapsulate entire body in the center of image
Clothing: Underwear
Pose - Front - Feet: Shoulder width apart
Pose - Front - Arms: Facing the floor at a 45 degree angle
Pose - Front - Hands: Flat with length of wrists facing the camera
Pose - Side - Arms: Flat against body not obstructing front or back outline
Pose - Side - Hands: Flat with width of wrists facing the camera

Steps For Processing Image

Filter Input (all steps)

To minimize errors, only necessary information should be used. The challenge of having too little or too much information helps to minimize errors. An added benefit of filtering large datasets is reduced computation requirements for processing.

Remove noise
Narrow field to region of interest

Extract Features & Information (one or a combination of all)

Color information - looking for changes in color or detect objects.
Blob extraction - detecting adjacent, similarly colored pixels.
Edges and corners - examining changes in brightness to identify borders or objects.
Pattern recognition and template matching - adding basic intelligence by matching features with the features of known objects.

Todo's

Familiar with scikit-image library, numpy, matplotlib
- Scikit-image modules: data, io, feature, filter
- Numpy: numpy-array, operators
- Matplotlib: show image, plot, axes
Remove noise in input image (front image, side image)
- Read image as grayscale image
- Remove salt and pepper noise
- Edge-preserving and noise reducing denoising filter
Remove background of image
- Finding edges of human in image by roberts (canny / sobel) fitler
- Skin detection
- Merge regions to find the human in image (skimage.segmentation.*)
Body parts recognition
- Using skeletonization method to get structure of human body
- Compare with human body region to get the body parts
- Label each body parts
Calculate the measurements of body parts
- Calculate the body parts by draw eclipse

Documents

https://drive.google.com/drive/u/0/folders/0B7rKJ3215ikuOU4xTzZPWm9FVU0

ankitv89/Image-Processor