imageTagging

This application was created as part of the interview process for a Software job.

This application provides REST endpoints for finding objects in images using the Imagga apis.

Running the App

First compile the jar file using maven ./mvnw install
Next, build the docker image. docker build -t behnke19/imagetagging .
Create a file called .env in the project directory that has two entries with your Imagga credentials.

imaggakey=yourapikey
imaggasecret=yoursecretkey

Finally, start the app with docker compose. This will start a mysql container along with the image tagging app.
docker-compose up -d

Using the APIs

GET /images

Returns a list of all the images in the database. And optional query param ?objects=cat,hat can be used to fetch only
images that contain the provided objects.

GET /images/{imageId}

Returns a specific image specified by its ID.

POST /images

Submit an image to the service and optionally run object detection on it. The payload should be a multipart form with the following fields.
label - A text label for the image. (Optional)
detectObjects - A boolean value. If true then object detection will be processed for the given image. (Optional)
url - A url that points to an image. (Either an url or file must be provided)
file - An image file to be processed. (Either an url or file must be provided)

This endpoint returns the image with a list of detected objects if object detection was run.