Note: This will only work on certain camera trap's footage, and was developed for a specific project
Install tesseract from here: https://tesseract-ocr.github.io/tessdoc/Home.html
Currrently, due to early development the path to the tesseract.exe file needs to be manually updated in camera_time_detect.py.
Replace the path on line 13 (pytesseract.pytesseract.tesseract_cmd = r'C:\\Program Files\\Tesseract-OCR\\tesseract.exe') with the path to your installation directory. Leave the r intact at the start of the string.
eg.
pytesseract.pytesseract.tesseract_cmd = r'C:\\Program Files\\Tesseract-OCR\\tesseract.exe'
changes to
pytesseract.pytesseract.tesseract_cmd = r'my_path_to_tesseract.exe'
Make sure that any backslashes (\) are either doubled or changed to forward slashes. (/)
You can set up the python enviroment using pip and the requirements.txt file. If you don't have python installed look at www.python.org and follow the instructions.
From a terminal(cmd or powershell should both work on Windows) navigate to the folder containing this repository and run pip install -r requirements.txt or pip3 install -r requirements.txt
Either run from the command line using:
python camera_time_detect.py "my_directory_containing_videos_path" "my_csv_path"
You can add an optional -d flag to indicate that you want to output as a human readable date stamp instead of a timestamp.
python camera_time_detect.py "my_directory_containing_videos_path" "my_csv_path" -d
Or use as a library by using:
import camera_time_detect
write_to_csv("my_csv_path", run_dir("my_dir_path"))