cs401r_w2016:lab7
Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
cs401r_w2016:lab7 [2016/02/18 05:19] admin [Grading:] |
cs401r_w2016:lab7 [2021/06/30 23:42] (current) |
||
|---|---|---|---|
| Line 6: | Line 6: | ||
| ====Pre-requisite:==== | ====Pre-requisite:==== | ||
| - | You must install the python ''skimage'' package. This can be installed via ''conda install skimage''. | + | You must install the python ''skimage'' package. This can be installed via ''conda install scikit-image''. |
| ---- | ---- | ||
| Line 13: | Line 13: | ||
| For this lab, you will implement the Kalman filter algorithm on two datasets. Your notebook should produce two visualizations of the observations, the true state, the estimated state, and your estimate of the variance of your state estimates. | For this lab, you will implement the Kalman filter algorithm on two datasets. Your notebook should produce two visualizations of the observations, the true state, the estimated state, and your estimate of the variance of your state estimates. | ||
| - | For the first dataset, you will implement a simple random accelerations model and use it to track a simple target. Your final visualization should look something like the following: | + | **For the first dataset,** you will implement a simple random accelerations model and use it to track a simple target. Your final visualization should look something like the following: |
| {{ :cs401r_w2016:lab7_kf.png?direct |}} | {{ :cs401r_w2016:lab7_kf.png?direct |}} | ||
| - | For the second dataset, you will use a series of '(x,y)' observations that were derived from a vision processing task. For this dataset, you will need to set the parameters of your filter to whatever best values you can, and then show a visualization of the result. The visualization of the result should consist of two parts: | + | **For the second dataset,** you will use a series of '(x,y)' observations that were derived from a vision processing task. For this dataset, you will need to set the parameters of your filter to whatever best values you can, and then show a visualization of the result. The visualization of the result should consist of two parts: |
| - First, a series of frame snapshots showing the estimated position and uncertainty | - First, a series of frame snapshots showing the estimated position and uncertainty | ||
| - Second, a plot of how the ball moved around in the field of view | - Second, a plot of how the ball moved around in the field of view | ||
| - | Examples of both are here: | + | Finally, you should clearly explain in your notebook why you selected the parameters you did. |
| + | |||
| + | Examples of both visualizations are here: | ||
| {{ :cs401r_w2016:lab7_ball.png?direct |}} | {{ :cs401r_w2016:lab7_ball.png?direct |}} | ||
| Line 43: | Line 45: | ||
| For the first dataset, you will perform simple target tracking using the random accelerations model. You should use the equations given in Lecture 17. The data can be found here: | For the first dataset, you will perform simple target tracking using the random accelerations model. You should use the equations given in Lecture 17. The data can be found here: | ||
| - | [[http://hatch.cs.byu.edu/courses/stat_ml/kfdata.mat|Simple target tracking data]] | + | [[https://www.dropbox.com/s/11fgx6e5atfv3zy/kfdata.mat?dl=0|Simple target tracking data]] |
| There are two arrays in this .mat file: ''data'' contains the actual (noisy) observations at each timestep, while ''true_data'' contains the true x,y coordinates at each timestep. Note that your kalman filter should only use the noisy observations; the true x,y coordinates are only given for visualization purposes. | There are two arrays in this .mat file: ''data'' contains the actual (noisy) observations at each timestep, while ''true_data'' contains the true x,y coordinates at each timestep. Note that your kalman filter should only use the noisy observations; the true x,y coordinates are only given for visualization purposes. | ||
| Line 90: | Line 92: | ||
| The dataset is derived from a sequence of images, which can be downloaded here: | The dataset is derived from a sequence of images, which can be downloaded here: | ||
| - | [[http://hatch.cs.byu.edu/courses/stat_ml/ball_data.mat|Ball data]] | + | [[https://www.dropbox.com/s/1x7jtg162upc3o9/ball_data.mat?dl=0|Ball data]] |
| The data is a sequence of images of a ball rolling across a camera's field of view. We'd like to implement a tracker for the ball, but there's no way we can cope with images in the framework of the Kalman filter. Instead, we'll do a brief pre-processing step: we'll use template matching to search for the ball in each frame of video, and record the best ''(x,y)'' position of the template match. | The data is a sequence of images of a ball rolling across a camera's field of view. We'd like to implement a tracker for the ball, but there's no way we can cope with images in the framework of the Kalman filter. Instead, we'll do a brief pre-processing step: we'll use template matching to search for the ball in each frame of video, and record the best ''(x,y)'' position of the template match. | ||
| Line 136: | Line 138: | ||
| # assumes that "mus" contains a list of Gaussian means, and that "covs" is a list of Gaussian covariances | # assumes that "mus" contains a list of Gaussian means, and that "covs" is a list of Gaussian covariances | ||
| + | # also assumes you are running in Jupyter Notebook | ||
| + | |||
| + | from IPython.display import clear_output | ||
| + | |||
| for t in range(0, data.shape[0]): | for t in range(0, data.shape[0]): | ||
| Line 152: | Line 158: | ||
| plt.xlim([1, 360]) | plt.xlim([1, 360]) | ||
| plt.ylim([243,1]) | plt.ylim([243,1]) | ||
| - | | + | clear_output(wait = True) |
| plt.pause(0.01) | plt.pause(0.01) | ||
| </code> | </code> | ||
cs401r_w2016/lab7.1455772787.txt.gz · Last modified: 2021/06/30 23:40 (external edit)
Page Tools
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
