用python实现 识别图片手写数字

要使用 Python 实现手写数字识别，最常用的方法是使用深度学习模型。以下是一个基于 TensorFlow 的示例代码，它使用卷积神经网络（Convolutional Neural Network, CNN）来实现手写数字识别：

首先，我们需要导入必要的库：

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
import numpy as np
import matplotlib.pyplot as plt

然后，我们需要加载 MNIST 数据集：

(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()

现在，我们可以先预处理数据，将输入数据归一化为 0 到 1 之间的值：

x_train = x_train.astype("float32") / 255.0
x_test = x_test.astype("float32") / 255.0

接下来，我们需要将标签进行 one-hot 编码：

y_train = keras.utils.to_categorical(y_train, 10)
y_test = keras.utils.to_categorical(y_test, 10)

现在，我们可以构建 CNN 模型：

model = keras.Sequential(
    [
        keras.Input(shape=(28, 28, 1)),
        layers.Conv2D(32, kernel_size=(3, 3), activation="relu"),
        layers.MaxPooling2D(pool_size=(2, 2)),
        layers.Conv2D(64, kernel_size=(3, 3), activation="relu"),
        layers.MaxPooling2D(pool_size=(2, 2)),
        layers.Flatten(),
        layers.Dropout(0.5),
        layers.Dense(10, activation="softmax"),
    ]
)

然后，我们可以编译模型并开始训练：

model.compile(loss="categorical_crossentropy", optimizer="adam", metrics=["accuracy"])
model.fit(x_train, y_train, batch_size=128, epochs=15, validation_split=0.1)

最后，我们可以使用测试集来评估模型的性能：

score = model.evaluate(x_test, y_test, verbose=0)
print("Test loss:", score[0])
print("Test accuracy:", score[1])

完整的代码如下：

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
import numpy as np
import matplotlib.pyplot as plt

# Load MNIST dataset
(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()

# Preprocess data
x_train = x_train.astype("float32") / 255.0
x_test = x_test.astype("float32") / 255.0

# One-hot encode labels
y_train = keras.utils.to_categorical(y_train, 10)
y_test = keras.utils.to_categorical(y_test, 10)

# Build CNN model
model = keras.Sequential(
    [
        keras.Input(shape=(28, 28, 1)),
        layers.Conv2D(32, kernel_size=(3, 3), activation="relu"),
        layers.MaxPooling2D(pool_size=(2, 2)),
        layers.Conv2D(64, kernel_size=(3, 3), activation="relu"),
        layers.MaxPooling2D(pool_size=(2, 2)),
        layers.Flatten(),
        layers.Dropout(0.5),
        layers.Dense(10, activation="softmax"),
    ]
)

# Compile model
model.compile(loss="categorical_crossentropy", optimizer="adam", metrics=["accuracy"])

# Train model
model.fit(x_train, y_train, batch_size=128, epochs=15, validation_split=0.1)

# Evaluate model
score = model.evaluate(x_test, y_test, verbose=0)
print("Test loss:", score[0])
print("Test accuracy:", score[1])

这个模型可以达到 99% 的准确率，可以用来识别手写数字。