A Flask interface to imagehub to assist the management of cameras and the images saved and catalogued by this application. This application provides access to the Object Detection feature, as well as an Automated license plate reader (ALPR) if you wish to activate this feature.
Contents
Select the camera you wish to display images.
This is an example screen of the Cameras for my home installation. Every installation of Image Librarian will contain
the Imagehub camera (DO NOT DELETE). This "camera" is the Camera Node that all "system events" are associated with
in the imagehub application.
A page will appear with all of the images for that camera in descending order. The number of images currently stored will appear in parenthesis next to camera name. Image may be clicked to enlarge, plus once enlarged the images may be examined in order by pressing the arrow keys on your keyboard or on the screen.
This will display the images from all the cameras in descending order.
Select a camera and update the parameters. If any field needs updating, click the Update button and edit the camera
data and Submit. Camera entries may also be Deleted if necessary.
The NodeName is a descriptive name for the camera. I choose to concatenate the node.name from the
imagenode.yaml file with the cameras.P1.viewname (e.g. 'BirdFeeder' and 'RPiCam3'). (REQUIRED)
The ViewName field should match the cameras.P1.viewname field in the imagenode.yaml file as seen below (REQUIRED)
# Settings for imagenode.py webcam motion detector testing---
node:
name: BirdFeeder # <-------
queuemax: 20
patience: 30
#heartbeat: 1
send_type: jpg
send_threading: True # sends images in separate thread
threaded_read: False # this is the new option; False selects PiCameraUnthreadedStream
stall_watcher: True # watches for stalled network or RPi power glitch
print_settings: True
hub_address:
H1: tcp://10.0.0.210:5555
cameras:
P1:
viewname: RPiCam3 # <-------
resolution: (800,600)
exposure_mode: sports
framerate: 30
detectors:
motion:
ROI: (5,25),(95,85)
draw_roi: ((255,0,0),1)
send_frames: detected event # continuous or none or detected event
send_count: 4 # number of images to send when an event occurs
delta_threshold: 5 # The minimum intensity difference between the current image and the weighted average of past images
min_motion_frames: 7 # The minimum number of frames with detected motion to change the state to "moving"
min_still_frames: 4 # The minimum number of frames with no detected motion to change the state to "still"
min_area: 2 # minimum area of motion as percent of ROI
blur_kernel_size: 17 # Guassian Blur kernel size - integer and odd
send_test_images: False
print_still_frames: False # default = True
draw_time: ((0,255,0),1)
draw_time_org: (5,5)
draw_time_fontScale: 0.6
This is a handy spot to document the type of camera used or information of importance to you. (OPTIONAL)
If checked or true, this will allow the camera images to be displayed in the desktop module dashboard.py. (OPTIONAL)
If checked or true, this will allow the camera images to be checked for objects by MQTT_client.py. (OPTIONAL)
If checked or true, this will allow the camera images to be checked by PLATE RECOGNIZER.
You MUST setup an account with PLATE RECOGNIZER, and receive an API Token to activate and use their services. (OPTIONAL)
This is name(s) found in cameras.P1.detectors.motion.roi_name field of the imagenode.yaml file located in the
RPi camera as seen below:
# Settings for imagenode.py webcam motion detector testing
---
node:
name: StreetView
queuemax: 50
patience: 15
heartbeat: 1
send_type: jpg
#send_threading: True # sends images in separate thread
#stall_watcher: True # watches for stalled network or RPi power glitch
print_settings: True
hub_address:
H1: tcp://10.0.0.100:5555
cameras:
P1:
viewname: RPiCam6
resolution: (800,600)
exposure_mode: auto
framerate: 30
detectors:
- motion:
ROI: (4,18),(70,48)
roi_name: Street # <-----------------
log_roi_name: False
draw_roi: ((0,255,0),1)
send_frames: detected event # continuous, none or detected event
send_count: 7 # number of images to send when an event occurs
delta_threshold: 7 # The minimum intensity difference between the current image and the weighted average of past images
min_motion_frames: 5 # The minimum number of frames with detected motion to change the state to "moving"
min_still_frames: 5 # The minimum number of frames with no detected motion to change the state to "still"
min_area: 3 # minimum area of motion as percent of ROI
blur_kernel_size: 21 # Guassian Blur kernel size - integer and odd
send_test_images: False
print_still_frames: False # default = True
draw_time: ((0,200,0),1)
draw_time_org: (5,5)
draw_time_fontScale: 0.5
- motion:
ROI: (28,50),(90,90)
roi_name: FrontDoor # <-----------------
log_roi_name: True
draw_roi: ((0,255,0),1)
send_frames: detected event # continuous, none or detected event
send_count: 7 # number of images to send when an event occurs
delta_threshold: 7 # The minimum intensity difference between the current image and the weighted average of past$
min_motion_frames: 5 # The minimum number of frames with detected motion to change the state to "moving"
min_still_frames: 5 # The minimum number of frames with no detected motion to change the state to "still"
min_area: 3 # minimum area of motion as percent of ROI
blur_kernel_size: 21 # Guassian Blur kernel size - integer and odd
send_test_images: False
print_still_frames: False # default = True
This is the name of the Region of Interest (ROI) to watch by the Image Librarian. For example, the imagenode.yaml
file above defines an ROI to watch for people approching the FrontDoor. Unfortunately, at this time only ONE ROI
can be defined per camera.
This is the Message used to send or notify you of a specific event.
If enabled, Text messages composed of the "Obect" + "Message" will be sent to the specified email address in email Google Voice
node of the Image Librarian Flow of Node-Red.
This option provides a look at the latest events for selected camera.
This option will display the last event to occur for each camera. This is handy for monitoring the last activity for all cameras at one time.
Select a License Plate of interest to see all the stored images for that plate.
Display all ALPR recorded events plus it provides a means of editing the ALPR event by clicking on the link below each image.
This option provides a means of Updating and Deleting a License Plate in the database. Great caution should
be taken in deleting entries in this Table since other Tables point to these entries.
Add a License Plate.
Enter the License Plate Number in Uppercase.
Enter the color of the vehicle.
Enter the type of vehicle (e.g. car, suv, sedan, pickup truck, big truck, etc.).
Enter any identifying information in this field.
Select an Object to view all images for this object.
View images and update Object data for images. Click the object link below the image to update object data.
Depricated.
Depricated. The Objects detected by this application are defined in the ~/IOTstack/tools/coco.names file. Many of
these are household items, and are unlikely to be detected in outdoor cameras.
