showing boundingboxes and label on my image detected by my trained model

hi everyone,

i am trinig to show label and boundingboxes by detecting by yolov2 trained model but error appera as and it can show any boxes on images

>> untitled3

Incorrect number or types of inputs or outputs for function 'detect'.

Error in untitled3 (line 4)

[bboxes,scores,labels] = detect(dnet,img,'Threshold',thrusholdvalue);

Answer 1

[Moksh]

Hi Ahmad,

I understand that you are using the “yolov2” pre-trained model to detect regions of interest, and subsequently, you aim to add bounding boxes around these detected regions.

While using the “detect” function two inputs need to be passed:

• Detector - A YOLO object detector.
• Image – Input image on which detection needs to be performed.

Further, the threshold can be passed using the “Threshold” parameter followed by a double value.

Here is an example code:

% Loading a pretrained yolov2 for vechicle detection
vehicleDetector = load('yolov2VehicleDetector.mat','detector');
detector = vehicleDetector.detector;

% Reading a sample image
I = imread('highway.png');
% imshow(I);

% Running the model on sample image for detection
[bboxes,scores,labels] = detect(detector,I,"Threshold",0.2);

% Plotting the bounding box on detected regions
if ~isempty(bboxes)
    detectedI = insertObjectAnnotation(I,'rectangle',bboxes,cellstr(labels));
end
figure
imshow(detectedI)

For more information about the “detect” function, please refer to the following documentation:

https://in.mathworks.com/help/vision/ref/yolov2objectdetector.detect.html

Hope this information helps!

Best Regards,

Moksh Aggarwal

[ahmad]

But my detector shows multiple boxes on same object.second think is that when I change threshold from 0.1 to 0?5 it doesn't detecte any thing.

[Image Analyst]

Well you accepted the answer so I guess youi got it figured out.