From bc6c115d453538b59eeb3b40371759e49aecd716 Mon Sep 17 00:00:00 2001
From: 2509165031 <2509165031@student.edu.cn>
Date: Thu, 16 Apr 2026 15:54:57 +0800
Subject: [PATCH] =?UTF-8?q?=E5=AE=8C=E6=88=90=E4=BD=9C=E4=B8=9A=E4=B8=80?=
 =?UTF-8?q?=EF=BC=9A3-1-3?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 0416 2.py | 32 ++++++++++++++++++++++++++++++++
 0416 3.py | 24 ++++++++++++++++++++++++
 0416.py   | 29 +++++++++++++++++++++++++++++
 3 files changed, 85 insertions(+)
 create mode 100644 0416 2.py
 create mode 100644 0416 3.py
 create mode 100644 0416.py

diff --git a/0416 2.py b/0416 2.py
new file mode 100644
index 0000000..dd70bc0
--- /dev/null
+++ b/0416 2.py	
@@ -0,0 +1,32 @@
+import numpy as np
+
+# 定义4×4图像矩阵
+img = np.array([
+    [255, 255, 0, 0],
+    [255, 255, 0, 0],
+    [0, 0, 255, 255],
+    [0, 0, 255, 255]
+], dtype=np.uint8)
+
+print("原始图像：")
+print(img)
+print("-" * 30)
+
+# 1. 统计白色(255)和黑色(0)像素数量
+white_count = np.sum(img == 255)
+black_count = np.sum(img == 0)
+print(f"白色像素(255)数量：{white_count}")
+print(f"黑色像素(0)数量：{black_count}")
+print("-" * 30)
+
+# 2. 水平翻转并打印
+img_flip = np.fliplr(img)
+print("水平翻转后图像：")
+print(img_flip)
+print("-" * 30)
+
+# 3. 逆时针旋转90度（转置+上下翻转）
+# 方法：先转置，再上下翻转(np.flipud)
+img_rot90_ccw = np.flipud(img.T)
+print("逆时针旋转90度后图像：")
+print(img_rot90_ccw)
\ No newline at end of file
diff --git a/0416 3.py b/0416 3.py
new file mode 100644
index 0000000..993b47d
--- /dev/null
+++ b/0416 3.py	
@@ -0,0 +1,24 @@
+import numpy as np
+
+# 假设这是从图像中提取的2个特征图
+feature_map1 = np.array([[1, 0, 1], [0, 1, 0], [1, 0, 1]])
+feature_map2 = np.array([[1, 1, 1], [1, 0, 0], [1, 0, 0]])
+
+# 1) 全代码：将特征图展平为向量
+vector1 = feature_map1.flatten()  # 展平
+vector2 = feature_map2.flatten()  # 展平
+
+print("vector1:", vector1)
+print("vector2:", vector2)
+print("-" * 40)
+
+# 1. 计算 vector1 和 vector2 的欧几里得距离
+euclidean_dist = np.linalg.norm(vector1 - vector2)
+print(f"欧几里得距离: {euclidean_dist:.4f}")
+
+# 2. 计算 vector1 和 vector2 的余弦相似度
+dot_product = np.dot(vector1, vector2)
+norm_a = np.linalg.norm(vector1)
+norm_b = np.linalg.norm(vector2)
+cos_similarity = dot_product / (norm_a * norm_b)
+print(f"余弦相似度: {cos_similarity:.4f}")
\ No newline at end of file
diff --git a/0416.py b/0416.py
new file mode 100644
index 0000000..690d3fa
--- /dev/null
+++ b/0416.py
@@ -0,0 +1,29 @@
+import numpy as np
+
+# 1. 定义原始3×3灰度图像
+image = np.array([
+    [100, 150, 200],
+    [80, 120, 180],
+    [60, 90, 140]
+], dtype=np.uint8)
+
+print("原图：")
+print(image)
+print("-" * 20)
+
+# 2. 变暗20：每个像素值减20（uint8会自动截断负数，这里所有值减20后均为正，无溢出）
+image_dark = image - 20
+print("变暗20后：")
+print(image_dark)
+print("-" * 20)
+
+# 3. 裁剪左上角2×2区域：image[0:2, 0:2]
+image_crop = image_dark[0:2, 0:2]
+print("裁剪左上角2×2后：")
+print(image_crop)
+print("-" * 20)
+
+# 4. 水平翻转：使用np.fliplr()
+image_flip = np.fliplr(image_crop)
+print("水平翻转后：")
+print(image_flip)
\ No newline at end of file