Wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
=== Installation ===
<code bash>
git clone https://github.com/ElektorLabs/CaptureCount.git
cd CaptureCount/yolo
wget https://pjreddie.com/media/files/yolov3.weights
sudo apt install python3-opencv
sudo apt install python3-numpy python3-pandas
</code>

=== Code ===

Largement inspiré de https://github.com/ElektorLabs/CaptureCount

<code python>
# coding: utf-8

# Working Object Tracker V1.2, adapted from https://www.learnopencv.com/deep-learning-based-object-detection-using-yolov3-with-opencv-python-c/
# This script will detect objects in a picture


import cv2
import pandas as pd
import time
from datetime import datetime
import numpy as np

# Load pre-trained model and configuration
model = './yolo/yolov3.weights' # Change this to your model's path
config = './yolo/yolov3.cfg'    # Change this to your config file's path
net = cv2.dnn.readNetFromDarknet(config, model)

# Load class names
classes = []
with open("./yolo/coco.names", "r") as f: # Change to the path of your coco.names file
    classes = [line.strip() for line in f.readlines()]

# Function to get output layers
def get_output_layers(net):
    layer_names = net.getLayerNames()
    return [layer_names[i - 1] for i in net.getUnconnectedOutLayers().flatten()]


# Function to draw bounding box
def draw_bounding_box(img, class_id, confidence, x, y, x_plus_w, y_plus_h):
    label = str(classes[class_id])
    color = np.random.uniform(0, 255, size=(3,))
    cv2.rectangle(img, (x, y), (x_plus_w, y_plus_h), color, 2)
    cv2.putText(img, label, (x-10, y-10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, color, 2)

frame = cv2.imread('chat.png')

Width = frame.shape[1]
Height = frame.shape[0]
scale = 0.00392

# Create a blob and pass it through the model
print("Creating a blob...")
blob = cv2.dnn.blobFromImage(frame, scale, (416, 416), (0, 0, 0), True, crop=False)
net.setInput(blob)
outs = net.forward(get_output_layers(net))
print("Blob created.")
# Process the outputs
class_ids = []
confidences = []
boxes = []
conf_threshold = 0.5
nms_threshold = 0.4

for out in outs:
    for detection in out:
        scores = detection[5:]
        class_id = np.argmax(scores)
        confidence = scores[class_id]
        if confidence > conf_threshold:
            center_x = int(detection[0] * Width)
            center_y = int(detection[1] * Height)
            w = int(detection[2] * Width)
            h = int(detection[3] * Height)
            x = center_x - w / 2
            y = center_y - h / 2
            class_ids.append(class_id)
            confidences.append(float(confidence))
            boxes.append([x, y, w, h])

indices = cv2.dnn.NMSBoxes(boxes, confidences, conf_threshold, nms_threshold)

for i in indices:
    box = boxes[i]
    x = box[0]
    y = box[1]
    w = box[2]
    h = box[3]
    draw_bounding_box(frame, class_ids[i], confidences[i], round(x), round(y), round(x+w), round(y+h))
    print("Box",i,"added.")

cv2.imwrite('chat.cv.png',frame)
</code>

=== Avant/Après ===

{{:les-gens-de-la-terrasse-d-un-cafe-paris-france-c1g5cy.jpg?direct&600 |}}

{{ :les-gens-de-la-terrasse-d-un-cafe-paris-france-c1g5cy.cv.jpg?direct&600|}}