import os
import numpy as np
import matplotlib.pyplot as plt
import re
from matplotlib.ticker import MaxNLocator
# 解决中文显示问题
plt.rcParams['font.sans-serif'] = ['SimHei', 'Microsoft YaHei', 'WenQuanYi Micro Hei']
plt.rcParams['axes.unicode_minus'] = False
def natural_sort_key(s):
"""自然排序算法:确保文件名按数字顺序排列"""
return [int(text) if text.isdigit() else text.lower()
for text in re.split(r'(\d+)', s)]
def find_stable_intervals(counts, min_window=300, max_window=2000,
std_threshold=10.0, merge_gap=300, min_length=500):
"""
改进版稳定区间检测:使用标准差作为稳定性指标
:param counts: 预测框数量列表
:param min_window: 最小窗口尺寸
:param max_window: 最大窗口尺寸
:param std_threshold: 标准差阈值(波动范围)
:param merge_gap: 相邻区间合并的最大间隔
:param min_length: 最小有效区间长度
:return: 优化后的稳定区间列表
"""
n = len(counts)
if n == 0:
return []
# 1. 自适应窗口机制
window_size = min(max_window, max(min_window, n // 10))
step_size = max(1, window_size // 2) # 50%重叠滑动
# 2. 初始检测稳定区间(使用标准差)
base_intervals = []
for i in range(0, n - window_size + 1, step_size):
window = counts[i:i + window_size]
if len(window) < 2: # 至少需要2个点计算标准差
continue
# 计算标准差作为稳定性指标
std_dev = np.std(window)
if std_dev < std_threshold:
base_intervals.append((i, i + window_size - 1))
# 如果没有检测到任何区间,直接返回
if not base_intervals:
return []
# 3. 合并相邻平稳段
base_intervals.sort(key=lambda x: x[0]) # 确保按起始索引排序
merged_intervals = []
current_start, current_end = base_intervals[0]
for start, end in base_intervals[1:]:
if start - current_end <= merge_gap: # 间隔小于合并阈值
current_end = max(current_end, end) # 扩展当前区间
else:
merged_intervals.append((current_start, current_end))
current_start, current_end = start, end
merged_intervals.append((current_start, current_end))
# 4. 过滤短时伪平稳段
final_intervals = [
(start, end) for start, end in merged_intervals
if (end - start + 1) >= min_length # 区间长度包含两端点
]
return final_intervals
def plot_box_count_trend_with_stable_intervals(file_list, box_counts, stable_intervals, output_path):
"""
绘制预测框数量变化趋势图并标记稳定区间
:param file_list: 文件名列表
:param box_counts: 预测框数量列表
:param stable_intervals: 稳定区间列表
:param output_path: 输出图片路径
"""
plt.figure(figsize=(20, 10))
# 绘制整体趋势
plt.plot(file_list, box_counts, 'b-', linewidth=1.5, label='预测框数量')
# 标记稳定区间
for i, (start, end) in enumerate(stable_intervals):
interval_files = file_list[start:end + 1]
interval_counts = box_counts[start:end + 1]
if not interval_counts: # 确保区间有效
continue
# 计算区间统计量
avg_count = np.mean(interval_counts)
min_count = np.min(interval_counts)
max_count = np.max(interval_counts)
std_dev = np.std(interval_counts)
# 绘制稳定区间
plt.fill_between(interval_files,
min_count, max_count,
color='green',
alpha=0.2,
label=f'稳定区间{i + 1}' if i == 0 else "")
# 添加区间标注
mid_idx = start + (end - start) // 2
plt.annotate(f"区间{i + 1}: {start + 1}-{end + 1}\n均值: {avg_count:.1f}±{std_dev:.1f}",
(file_list[mid_idx], avg_count),
xytext=(0, 20),
textcoords='offset points',
ha='center',
fontsize=10,
bbox=dict(boxstyle="round,pad=0.3", fc="yellow", alpha=0.7))
# 设置图表属性
plt.title('预测框数量变化趋势及稳定区间分析', fontsize=18)
plt.xlabel('图像文件名', fontsize=14)
plt.ylabel('预测框数量', fontsize=14)
plt.xticks(rotation=90, fontsize=7)
plt.grid(True, linestyle='--', alpha=0.6)
plt.legend(loc='upper right')
# 添加统计信息
stats_text = f"总文件数: {len(file_list)}\n稳定区间数: {len(stable_intervals)}"
plt.figtext(0.95, 0.95, stats_text,
ha='right', va='top',
bbox=dict(facecolor='white', alpha=0.8),
fontsize=12)
# 限制X轴刻度数量
plt.gca().xaxis.set_major_locator(MaxNLocator(20))
plt.tight_layout()
plt.savefig(output_path, dpi=150, bbox_inches='tight')
plt.close()
# 配置路径
label_dir = "D:/630-3-label-combine" # 替换为您的标签文件夹路径
output_dir = "D:/630-report" # 输出目录
os.makedirs(output_dir, exist_ok=True)
# 获取文件列表并按自然顺序排序
file_list = [f for f in os.listdir(label_dir) if f.endswith(".txt")]
file_list.sort(key=natural_sort_key)
# 提取文件名(不含扩展名)
file_names = [os.path.splitext(f)[0] for f in file_list]
# 统计每个文件的预测框数量
box_counts = []
for file in file_list:
file_path = os.path.join(label_dir, file)
count = 0
with open(file_path, 'r') as f:
for line in f:
if line.strip(): # 非空行
count += 1
box_counts.append(count)
# 计算整体统计数据
total_mean = np.mean(box_counts)
total_std = np.std(box_counts)
# 找出稳定区间(使用标准差作为指标)
stable_intervals = find_stable_intervals(
box_counts,
min_window=300, # 最小检测窗口
max_window=2000, # 最大检测窗口
std_threshold=total_std * 0.5, # 基于整体标准差设置阈值
merge_gap=300, # 合并最大间隔
min_length=500 # 最小有效长度
)
# 生成结果图片
output_path = os.path.join(output_dir, "box_count_stable_intervals_std.png")
plot_box_count_trend_with_stable_intervals(file_names, box_counts, stable_intervals, output_path)
# 输出详细结果
print(f"分析完成! 共处理 {len(file_list)} 个文件")
print(f"整体平均框数: {total_mean:.2f} ± {total_std:.2f}")
print(f"发现 {len(stable_intervals)} 个稳定区间:")
for i, (start, end) in enumerate(stable_intervals):
interval_counts = box_counts[start:end + 1]
avg_count = np.mean(interval_counts)
std_dev = np.std(interval_counts)
print(f"区间{i + 1}:")
print(f" - 文件范围: {start + 1}-{end + 1} (共{end - start + 1}个文件)")
print(f" - 平均框数: {avg_count:.2f} ± {std_dev:.2f}")
print(f" - 最小值: {min(interval_counts)}, 最大值: {max(interval_counts)}")
print(f"结果图片已保存至: {output_path}")
# 保存区间信息到文本文件
interval_info_path = os.path.join(output_dir, "stable_intervals_report_std.txt")
with open(interval_info_path, 'w') as f:
f.write(f"稳定区间分析报告(标准差指标)\n")
f.write(f"总文件数: {len(file_list)}\n")
f.write(f"整体平均框数: {total_mean:.2f} ± {total_std:.2f}\n")
f.write(f"稳定区间数: {len(stable_intervals)}\n\n")
for i, (start, end) in enumerate(stable_intervals):
interval_counts = box_counts[start:end + 1]
avg_count = np.mean(interval_counts)
std_dev = np.std(interval_counts)
f.write(f"区间 {i + 1}:\n")
f.write(f" 起始文件索引: {start + 1} ({file_names[start]})\n")
f.write(f" 结束文件索引: {end + 1} ({file_names[end]})\n")
f.write(f" 文件数量: {end - start + 1}\n")
f.write(f" 平均预测框数: {avg_count:.2f} ± {std_dev:.2f}\n")
f.write(f" 最小值: {min(interval_counts)}, 最大值: {max(interval_counts)}\n")
f.write("-" * 50 + "\n")
print(f"详细区间报告已保存至: {interval_info_path}")
将代码中的标准差改为标准差和变异系数和趋势斜率,分别生成三张图,再将三个指标生成的稳定区间合一起生成一张图