Exif pilot tutorial install#
Exif pilot tutorial code#

# if none of these exist, then use current timestampĮs_doc = str ( datetime. # get datetime meta data from one of these keys if possibleĮs_doc = exif_data # call the method to have PIL return exif dataĮs_doc = Image. Print ( 'create_exif_data PIL ERROR:', error, '- for file:', _file )

# create PIL Image from path and file name # put PIL Image conversion in a try-except indent block # create a new dict obj for the Elasticsearch doc Then this script will 'fill in the gaps' for the new documents If certain meta data is missing from the orginal, # if img file doesn't have _getexif, then give empty dictįunction to create new meta data for the Elasticsearchĭocument. Print ( 'get_image_meta AttributeError for:', file_name, '-', error ) # call the img's getexif() method and return EXIF data # use PIL to verify image is not corrupted # create a client instance of ElasticsearchĮlastic_client = Elasticsearch ( )įunction that uses PIL's TAGS class to get an image's EXIF

Exif pilot tutorial code#

Once you’ve confirmed this, we’ll add the following code to create a new client instance of Elasticsearch in the Python script: Create a new client instance of the Elasticsearch low-level clientīefore we proceed, let’s make sure that the Elasticsearch cluster is up and running. Pillow (commonly known as PIL), elasticsearch, and pybase64 (or base64 for Python) are the only libraries we’ll be using in this article that don’t come packaged with a Python installation by default. If you get an ImportError, then return to the Prerequisites section of this article and make sure all of the libraries are installed and updated. Import datetime # for image EXIF data timestampĪfter making these edits, save your script and run it in a terminal or command prompt window (using the python3 command) to make sure that all modules were properly installed using pip3. Import base64 # convert image to b64 for indexing # import the Image and TAGS classes from Pillow (PIL) # import the Elasticsearch low-level client

Exif pilot tutorial install#

Introduction Prerequisites for indexing photos to Elasticsearch Install all the Python libraries and dependencies using pip3 Install pybase64 for the Base64 Python encoding library Select a personal photo or public domain image to index to Elasticsearch Create a new project directory and Python script Import the necessary Python libraries and modules to index an image to Elasticsearch Create a new client instance of the Elasticsearch low-level client Start the Kibana service Use Python’s PIL library to grab the image’s EXIF meta data Set up the image document’s index data for Elasticsearch Create a PIL Image instance of the image in the directory Use’s PIL’s PIL.ExifTags Python library to read an image’s EXIF meta data Create a Python function to get the PIL Image’s EXIF data Create another function to generate missing EXIF data if needed Create a NumPy array Convert the Python list object to a str Use Python’s Base64 library to encode images Put the raw image data into the _source object Create an Elasticsearch index and put the image data into it Create an Elasticsearch index and ignore any 400 HTTP errors Index the image dict object to the Elasticsearch index Getting the raise ConnectionTimeout(“TIMEOUT”, str(e), e) error while calling the Elasticsearch index() method Conclusion Open up the Kibana Console UI to verify that the image was indexed Just the Code