Image Data ML Model Development
Budget: ₹1,500 – ₹12,500 INR
The main aim of this task is to Develop a model that uses satellite
signal characteristics — such as SNR (Signal-to-Noise Ratio), elevation angle, and azimuth — to
classify whether the signal is LOS or NLOS. This task involves the following sub-tasks but not
limited to:
1. Preprocessing of the GNSS data given
2. Features extraction from the Given Signal Data
3. Modeling the classifier
4. Performance evaluation through confusion matrix for label LOS and NLOS
Input Data: the input data is given in the image format
Output Data: An excel sheet or CSV sheet which includes confusion matrix as shown below and performance metrics like Recall, Precision, F-score and Accuracy;
SKILLLS REQUIRED:-
To successfully complete the GNSS LOS/NLOS classification project, a combination of programming, data processing, and machine learning skills is essential. Proficiency in Python programming is required to handle data preprocessing, file operations, and scripting the entire workflow. Since the input data is in image format, familiarity with image processing libraries like OpenCV or PIL is important for tasks such as resizing, grayscale conversion, and noise reduction. Knowledge of machine learning and deep learning, particularly with convolutional neural networks (CNNs) using frameworks like TensorFlow or Keras, is crucial for training and evaluating the classification model.
In addition, skills in data preprocessing—such as normalization and train-test splitting—are necessary to prepare the dataset for modeling. Understanding key evaluation metrics like confusion matrix, precision, recall, accuracy, and F1-score is important for assessing the model's performance. The ability to use pandas and Excel libraries helps in saving results and creating deliverables in CSV or Excel formats.
Furthermore, a basic understanding of GNSS concepts like Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) and how they affect satellite signal quality helps in better interpreting the data. Strong analytical thinking, problem-solving, and debugging skills are also valuable throughout the project.
signal characteristics — such as SNR (Signal-to-Noise Ratio), elevation angle, and azimuth — to
classify whether the signal is LOS or NLOS. This task involves the following sub-tasks but not
limited to:
1. Preprocessing of the GNSS data given
2. Features extraction from the Given Signal Data
3. Modeling the classifier
4. Performance evaluation through confusion matrix for label LOS and NLOS
Input Data: the input data is given in the image format
Output Data: An excel sheet or CSV sheet which includes confusion matrix as shown below and performance metrics like Recall, Precision, F-score and Accuracy;
SKILLLS REQUIRED:-
To successfully complete the GNSS LOS/NLOS classification project, a combination of programming, data processing, and machine learning skills is essential. Proficiency in Python programming is required to handle data preprocessing, file operations, and scripting the entire workflow. Since the input data is in image format, familiarity with image processing libraries like OpenCV or PIL is important for tasks such as resizing, grayscale conversion, and noise reduction. Knowledge of machine learning and deep learning, particularly with convolutional neural networks (CNNs) using frameworks like TensorFlow or Keras, is crucial for training and evaluating the classification model.
In addition, skills in data preprocessing—such as normalization and train-test splitting—are necessary to prepare the dataset for modeling. Understanding key evaluation metrics like confusion matrix, precision, recall, accuracy, and F1-score is important for assessing the model's performance. The ability to use pandas and Excel libraries helps in saving results and creating deliverables in CSV or Excel formats.
Furthermore, a basic understanding of GNSS concepts like Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) and how they affect satellite signal quality helps in better interpreting the data. Strong analytical thinking, problem-solving, and debugging skills are also valuable throughout the project.