Tree Species Detection from Aerial Imagery
Budget: ₹12,500 – ₹37,500 INR
I am in need of a proficient freelancer with a strong background in image processing and machine learning. The goal is to develop a solution for identifying specific tree species (Neem, Karanja, Mahua) from drone-acquired aerial imagery. The task will involve analyzing high-resolution orthoimages and creating a model for accurate species identification.
Key Responsibilities:
- Process drone-acquired RGB imagery to detect the specified tree species.
- Implement and train Convolutional Neural Networks (CNNs) for species classification.
- Provide precise geotagging for the identified trees.
- Collaborate with my team to ensure smooth data integration.
- Develop a pipeline for processing and analyzing aerial imagery data.
- Evaluate and optimize the accuracy and performance of the classification model.
- Implement image pre-processing steps such as normalization, augmentation, and resizing.
- Document the workflow, methodologies, and findings in a comprehensive report.
Ideal Skills and Experience:
- Extensive experience with image processing and machine learning.
- Proficient in using Convolutional Neural Networks (CNNs).
- Familiar with processing and analyzing drone-acquired RGB imagery.
- Excellent collaboration skills for seamless teamwork.
The identified tree species data should be provided in GeoJSON format. Using PyTorch for model training and evaluation is preferred. The species identification model should aim for an accuracy between 80% and 90%. The project should be completed within 1-3 months. Geotagging should be moderately precise, balancing detail and resource constraints.
Key Responsibilities:
- Process drone-acquired RGB imagery to detect the specified tree species.
- Implement and train Convolutional Neural Networks (CNNs) for species classification.
- Provide precise geotagging for the identified trees.
- Collaborate with my team to ensure smooth data integration.
- Develop a pipeline for processing and analyzing aerial imagery data.
- Evaluate and optimize the accuracy and performance of the classification model.
- Implement image pre-processing steps such as normalization, augmentation, and resizing.
- Document the workflow, methodologies, and findings in a comprehensive report.
Ideal Skills and Experience:
- Extensive experience with image processing and machine learning.
- Proficient in using Convolutional Neural Networks (CNNs).
- Familiar with processing and analyzing drone-acquired RGB imagery.
- Excellent collaboration skills for seamless teamwork.
The identified tree species data should be provided in GeoJSON format. Using PyTorch for model training and evaluation is preferred. The species identification model should aim for an accuracy between 80% and 90%. The project should be completed within 1-3 months. Geotagging should be moderately precise, balancing detail and resource constraints.