-
Notifications
You must be signed in to change notification settings - Fork 2
The goal of the project was to make a low cost and portable solution for liquid level identification using a combination of GoPro Hero 5 + Raspberry Pi 3 and Intel NCS. The system uses a combination of OpenCV edge detection and Caffe LeNet to identify the liquid level.
roborags/Level_Det_CaffeCNN_NCS
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
Level Detection using Caffe CNN on NCS: Project by Raghavan Naresh Sarangangapani : https://www.linkedin.com/in/roborags Done at the IUPUI Internet of Things Collaboratory, Indiana University Purdue University Indianapolis Under the guidance of Dr. Mohammed El-Sharkawy Demo video : https://youtu.be/FpKw6p6fX2Y Project details: 1. For main image dump and other processed image datasets, download the 7z files from following link and uncompress: https://www.dropbox.com/sh/8xfywmfk70nnnfh/AAAtJhhw-4mJsz4IhNf0wXx1a?dl=0 2. Project_Source_Code has all files for the project, read internal ReadMe to understand functions of each code. 3. Caffe files has the required Caffe CNN files. 4. Caffe model and solver state are compressed and split with 7z compression to accomodate in GitHub. 5. To install Caffe follow steps as in Caffe install steps document by Durvesh Pathak of IUPUI. 6. To install NCS packages refer report documentation. 7. Components used : a. Intel Neural Compute Stick and NCSDK b. GoPro Hero 4 Camera and GoPro Py API's c. Raspberry Pi 3 d. Test-tube setup with external lights e. GPU powered PC for training of Caffe network weights
About
The goal of the project was to make a low cost and portable solution for liquid level identification using a combination of GoPro Hero 5 + Raspberry Pi 3 and Intel NCS. The system uses a combination of OpenCV edge detection and Caffe LeNet to identify the liquid level.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published