LeafAreaCalculator
Updated. ComputerVison and Machine Learning: Convolutional Neural Network to compute leaf area
This project was a free lancing project for School of Biological Science at Washington State University. The goal of the project was to detect number of pixels that would potentially consitute a leaf. These 'green' pixels fraction as compared to a known quantity computes the area (cm^2). This constitues to a classification problem over image object.
Steps:
- Train CNN model to learn various levels of green, red and arbitrary pixels. [Note any other color except green and red is of no particular interest to us, hence classified as arbitrary.
- Break the image as list of pixels and classify them in a supervised training fashion. [Select pure green leaf images placed in 'green' folder, pure 'red' images in 'red' folder, etc]
- Compile and save the model that tests with an acceptable accuracy. [Accuracy obtained: 95 %].
- Knowing that the number of pixels in every image is 4 cm^2, we computed the leaf area as number of green pixels ~ number of red pixels [whose areas is known].
The project was implemeneted using Python Keras/Tensorflow libraries and OpenCV/color packages. A substantial advantage of this program was, due to the state of the art CNN machine learning algorithm used, leaf object with varied intensities of green were classfied with supreme accuracy. The raw images were generated through different sources, scanner and DSLR camera, but due to the algorithm logic, these differences were incorporated and computation did not result in loss of records.
This algorithm has been successfully run on atleast 3500+ images and has helped the research lab save valuable time and energy otherwise invested in manual efforts.
1 maintainer
Metadata
1 maintainer
Release files for pyleafarea 2.3.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| pyleafarea-2.3.1.tar.gz | 28.0 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| pyleafarea-2.3.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 56.0 MB
Release files / pyleafarea-2.3.1.tar.gz
| Download URL | pyleafarea-2.3.1.tar.gz |
|---|---|
| Size | 28.0 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
4c10b27f5cbbd4ac750515d0f2eb5ed8658bb31b92e0e0bd8bbf1538f723ece5
|
|
BLAKE2b-256 checksum How to use checksums |
40aeacde2124d19143c38b815bd02680ee6efaaf08498d9f2b7de3be85f4c180
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / pyleafarea-2.3.1-py3-none-any.whl
| Download URL | pyleafarea-2.3.1-py3-none-any.whl |
|---|---|
| Size | 28.0 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
eb115f575a8bade7302d14d07e157eec380f67953ebe92f4096a6ef33e577282
|
|
BLAKE2b-256 checksum How to use checksums |
bc307e29d6eacfa761f3711b19f4266cc513ed46b337353a8fc8444f6f3fc160
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
