基于opencv和knn的摄像头数字识别简单代码

这是一个基于OpenCV和k-NN算法的摄像头数字识别算法的简单代码：

import cv2
import numpy as np

# Load the data for the k-NN classifier
with np.load('knn_data.npz') as data:
    train = data['train']
    train_labels = data['train_labels']

# Setup the k-NN classifier
knn = cv2.ml.KNearest_create()
knn.train(train, cv2.ml.ROW_SAMPLE, train_labels)

# Setup the webcam
cap = cv2.VideoCapture(0)

while True:
    # Grab a frame from the webcam
    ret, frame = cap.read()

    # Convert the frame to grayscale
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    # Apply some preprocessing (thresholding and morphological closing)
    _, gray = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
    kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3, 3))
    gray = cv2.morphologyEx(gray, cv2.MORPH_CLOSE, kernel)

    # Find contours of digits in the frame
    _, contours, _ = cv2.findContours(gray, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

    # Loop over each contour
    for contour in contours:
        # Compute the bounding rectangle of the contour
        x, y, w, h = cv2.boundingRect(contour)

        # Ignore small contours
        if w < 15 or h < 15:
            continue

        # Extract the digit from the contour and resize it
        digit = gray[y:y+h, x:x+w]
        digit = cv2.resize(digit, (28, 28), interpolation=cv2.INTER_AREA)

        # Flatten the digit and convert it to a 1D array
        digit = digit.flatten()
        digit = np.float32(digit)

        # Classify the digit using the k-NN classifier
        _, result, _, _ = knn.findNearest(digit.reshape(1, -1), k=3)

        # Draw the classification result on the frame
        cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
        cv2.putText(frame, str(int(result[0])), (x, y), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)

    # Show the current frame
    cv2.imshow('frame', frame)

    # Exit if the user presses 'q'
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

# Release the webcam and destroy all windows
cap.release()
cv2.destroyAllWindows()

在这个代码中，k-NN分类器被用来识别出数字。训练数据和标签被存储在文件knn_data.npz中。每帧从网络摄像头中得到，然后用OpenCV的函数处理，以提取出图像中的数字。对于每个数字，应用相同的预处理步骤并将其输入k-NN分类器。分类器输出一个数字，我们将它作为识别结果画在帧上并用imshow函数显示在屏幕上。