Jetson Nano Jetbot Camera Image not showing in Object recognition

I’m following jetbot tutorials and trying to get Object Following Notebook example working, using both a standard Raspberry Pi and Waveshare Raspberry Pi Camera Module (G) w/ Fisheye Lens. While no errors are displayed, the image does not show running :

from jetbot import ObjectDetector
model = ObjectDetector(‘ssd_mobilenet_v2_coco.engine’)

from jetbot import Camera
camera = Camera.instance(width=300, height=300)

detections = model(camera.value)
print(detections)

from IPython.display import display
import ipywidgets.widgets as widgets
detections_widget = widgets.Textarea()
detections_widget.value = str(detections)
display(detections_widget)

image_number = 0
object_number = 0
print(detections[image_number][object_number])

dmesg shows:
jetbot@jetbot:~$ dmesg | grep camera
[ 0.544909] GPIO line 151 (camera-control-output-low) hogged as output/low
[ 0.748556] tegra_camera_platform tegra-camera-platform: tegra_camera_probe:camera_platform_driver probe
[ 0.748812] misc tegra_camera_ctrl: tegra_camera_isomgr_register isp_iso_bw=1500000, vi_iso_bw=750000, max_bw=1500000

though I get the same display with either Camera or the cable unplugged. So I guess my camera is not connected. What steps to follow?

Is there a simple way, like the Raspberry Pi command “raspistill -v -o test.jpg”, to check the Camera installation?

ANSWERED: I located on [url]https://github.com/JetsonHacksNano/CSI-Camera[/url] a very useful utility that checked the camera and face detection functions.

FIXED: While I thought my Pi and the Waveshare Cameras were V2 as required by the Nano, careful checking showed they were v1.3. Installing a new V2.1 worked with the CSI-Camera utility.
Thanks to JetsonHacksNano

1 Like