clean_annotation_files.py

from os import listdir, rename
from os.path import isfile, join
import pandas as pd
import os
from shutil import copy2

#mypath = 'C:/Users/cksas/Desktop/Annotation'
mypath = os.getcwd()
mypath = join(mypath, 'Actual')
resultsPath = os.getcwd()
resultsPath = join(resultsPath, 'CleanData')
folders = ['CAM1', 'CAM2', 'CAM3']
fl = pd.read_csv((join(mypath, 'FileList.csv')), names = folders)

for folder in folders:
    for i in range(10):
        name = fl[folder][i].split('.')[0]
        name = name + '.csv'
        newname = folder + '-' + str(i) + '_cleaned.csv'
        print (name)
        print (join(resultsPath, newname))
        copy2(join(mypath, folder, name), join(resultsPath, newname))

# Clean the data files
n1 = 'CAM1-'
n2 = 'CAM2-'
n3 = 'CAM3-'

for i in range(10):
    fnames = [n1 + str(i) + '_cleaned.csv', n2 + str(i) + '_cleaned.csv', n3 + str(i) + '_cleaned.csv']
    df1 = pd.read_csv(join(resultsPath, fnames[0]))
    df1 = df1.dropna()
    df2 = pd.read_csv(join(resultsPath, fnames[1]))
    df2 = df2.dropna()
    df3 = pd.read_csv(join(resultsPath, fnames[2]))
    df3 = df3.dropna()

    f2 = df2['frame']
    f3 = df3['frame']
    commonFrames = []

    for index, row in df1.iterrows():
        frameNum = row['frame']
        if frameNum in f2 and frameNum in f3:
            commonFrames.append(frameNum)

    print (len(commonFrames))

    df1 = df1.query('frame in @commonFrames')
    df2 = df2.query('frame in @commonFrames')
    df3 = df3.query('frame in @commonFrames')

    df1.to_csv(join(resultsPath, fnames[0].split('.')[0] + '.csv'))
    df2.to_csv(join(resultsPath, fnames[1].split('.')[0] + '.csv'))
    df3.to_csv(join(resultsPath, fnames[2].split('.')[0] + '.csv'))