I am teaching Remote Sensing and GIS, I would like to learn how to implement AI in image classification in the easiest way for geological mapping. My knowledge of AI is limited. Any suggestions or courses are appreciated.
Thank a lot.
Mohamed Matsah
ERDAS and ENVI are primarily designed for remote sensing and geospatial image analysis, and are not specifically designed for implementing AI-based image classification algorithms. However, they do have some capabilities for image classification, including traditional machine learning techniques such as decision trees and support vector machines.
That being said, if you want to implement AI-based image classification, there are other software tools that are more suitable for this purpose, such as TensorFlow, PyTorch, and Keras. These tools have easy-to-use APIs and pre-trained models that can be used for image classification tasks.
One of the easiest ways to implement AI in image classification is by using pre-trained models available in these frameworks. These models have already been trained on large datasets and can be fine-tuned for specific image classification tasks.
For instance, you can use transfer learning to fine-tune a pre-trained model like VGG, ResNet or Inception for your specific image classification problem. This involves replacing the last layer(s) of the pre-trained model with a new set of layers that are specifically designed for your classification task.
Once you have fine-tuned the model, you can use it to classify new images. This can be done using a simple Python script that loads the pre-trained model and passes the new images through the model for classification.
In summary, while ERDAS and ENVI have some image classification capabilities, if you want to implement AI-based image classification, it is best to use other software tools that are specifically designed for this purpose.
Dear Mohamed Matsah ,
Based on the widely-used commercial software package ERDAS IMAGINE 9.0, we developed an automated ANN classification system to the most commonly used image classifiers such as.
https://www.mdpi.com/2072-4292/1/3/243
Regards,
Shafagat
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence