Project Movie¶

The Project Movie calculates a frame statistic from a microscope movie, such as the mean frame.

This may be applied to a preprocessed movie in order to generate a reference frame for the Motion Correct algorithm. Alternatively, the statistic frame may be useful as a summary of the movie data at a particular processing stage.

Inputs¶

File Inputs¶

Description¶

Each pixel in the statistic frame or image of a movie can be calculated individually. For example, let's consider calculating the pixel value at coordinates \((x, y)\) of the mean frame:

where \(M_{t}\) is frame \(t\) of a movie \(M\), \(t_\text{from} = 0\), and \(t_\text{to} = T\) where \(T\) is the total number of frames in the movie. Any invalid frames, like dropped frames, are ignored in the calculation. This calculation is performed for each pixel in the frame to generate the final result. The minimum and maximum statistic image are calculated in the same way.

If you select local correlation as Statistic Type, Project Movie will compute the local correlation image the following way:

• for each pixel (pixel of interest, POI) in the field of view (FoV), this tool will load the POI's trace across time along with traces from its (up to) 8 neighboring pixels
• it will then compute correlation coefficients between the POI and each of its neighbors
• the POI will take as value the maximum correlation coefficient
• this process is repeated for all pixels of the FoV, to compose the local correlation image

This projection image is especially useful as a template image in the CaImAn Multi-Session Registration tool.