DOODS - Distributed Outside Object Detection Service

The doods image processing platform allows you to detect and recognize objects in a camera image using DOODS. The state of the entity is the number of objects detected, and recognized objects are listed in the summary attribute along with quantity. The matches attribute provides the confidence score for recognition and the bounding box of the object for each detection category.

Setup

You need to have DOODS running somewhere. It’s easiest to run as a docker container and deployment is described on docker hub DOODS - Docker

Configuration

The configuration loosely follows the tensorflow configuration. To enable this platform in your installation, add the following to your configuration.yaml file:

# Example configuration.yaml entry
image_processing:
  - platform: doods
    url: "http://<my doods server>:8080"
    detector: default
    source:
      - entity_id: camera.front_yard

Configuration Variables

source

(map)(Required)

The list of image sources.

entity_id

(string)(Required)

A camera entity id to get picture from.

name

(string)(Optional)

This parameter allows you to override the name of your image_processing entity.

url

(string)(Required)

The URL of the DOODS server

timeout

(integer)(Optional)

Timeout for requests (in seconds)

Default value:

detector

(string)(Required)

The DOODS detector to use

confidence

(float)(Optional)

The default confidence for any detected objects where not explicitly set

area

(map)(Optional)

Global detection area. Objects in this box will be reported. Top of image is 0, bottom is 1. Same left to right.

top

(float)(Optional)

Top line defined as % from top of image.

Default value:

left

(float)(Optional)

Left line defined as % from left of image.

Default value:

bottom

(float)(Optional)

Bottom line defined as % from top of image.

Default value:

right

(float)(Optional)

Right line defined as % from left of image.

Default value:

covers

(boolean)(Optional)

If true the detection must be fully in this box. If false any part of the detection in the box will trigger.

Default value:

true

file_out

(list)(Optional)

A template for the integration to save processed images including bounding boxes. camera_entity is available as the entity_id string of the triggered source camera.

labels

(map)(Optional)

Information about the selected labels model.

name

(string)(Required)

The label of the object to select for detection.

confidence

(float)(Optional)

The minimum confidence for the selected label

area

(map)(Optional)

Custom detection area. Only objects fully in this box will be reported. Top of image is 0, bottom is 1. Same left to right.

top

(float)(Optional)

Top line defined as % from top of image.

Default value:

left

(float)(Optional)

Left line defined as % from left of image.

Default value:

bottom

(float)(Optional)

Bottom line defined as % from top of image.

Default value:

right

(float)(Optional)

Right line defined as % from left of image.

Default value:

covers

(boolean)(Optional)

If true the detection must be fully in this box. If false any part of the detection in the box will trigger.

Default value:

true

# Example advanced configuration.yaml entry
# Example configuration.yaml entry
image_processing:
  - platform: doods
    scan_interval: 1000
    url: "http://<my doods server>:8080"
    timeout: 60
    detector: default
    source:
      - entity_id: camera.front_yard
    file_out:
      - "/tmp/{{ camera_entity.split('.')[1] }}_latest.jpg"
      - "/tmp/{{ camera_entity.split('.')[1] }}_{{ now().strftime('%Y%m%d_%H%M%S') }}.jpg"
    confidence: 50
    # This global detection area is required for all labels
    area:
      # Exclude top 10% of image
      top: 0.1
      # Exclude right 5% of image
      right: 0.95
      # The entire detection must be inside this box
      covers: true
    labels:
      - name: person
        confidence: 40
        area:
          # Exclude top 10% of image
          top: 0.1
          # Exclude right 15% of image
          right: 0.85
          # Any part of the detection inside this area will trigger
          covers: false
      - car
      - truck

Optimizing resources

Image processing components process the image from a camera at a fixed period given by the scan_interval. This leads to excessive processing if the image on the camera hasn’t changed, as the default scan_interval is 10 seconds. You can override this by adding to your config scan_interval: 10000 (setting the interval to 10,000 seconds), and then call the image_processing.scan service when you actually want to perform processing.

# Example advanced configuration.yaml entry
image_processing:
  - platform: doods
    scan_interval: 10000
    source:
      - entity_id: camera.driveway
      - entity_id: camera.backyard

# Example advanced automations.yaml entry
- alias: Doods scanning
  trigger:
     - platform: state
       entity_id:
         - binary_sensor.driveway
  action:
    - service: image_processing.scan
      entity_id: image_processing.doods_camera_driveway