Upload 3 files

app.py

@@ -9,8 +9,7 @@ from sort import * #运动检测，采用sort算法
 import tempfile #创建输出临时文件夹
 import os
 from detectMotion import * #单独的运动检测
-import matplotlib
-matplotlib.use('Agg')
+
 
 #导入YoloV8模型,第一次使用会下载模型到当前文件夹当中
 model=YOLO("yolov8n.pt")
@@ -34,6 +33,7 @@ tracker=Sort(max_age=20,min_hits=3,iou_threshold=0.3)
 
 # 彩色图像进行自适应直方图均衡化
 def hisEqulColor(img):
+    fig_preprocessed = image_histogram(img)
     ## 将RGB图像转换到YCrCb空间中
     ycrcb = cv2.cvtColor(img, cv2.COLOR_BGR2YCR_CB)
     # 将YCrCb图像通道分离
@@ -44,7 +44,9 @@ def hisEqulColor(img):
     clahe.apply(channels[0], channels[0])
     cv2.merge(channels, ycrcb)
     cv2.cvtColor(ycrcb, cv2.COLOR_YCR_CB2BGR, img)
-    return img
+    fig_postprocessed = image_histogram(img)
+    
+    return img, fig_preprocessed, fig_postprocessed
 
 #添加高斯噪声，并使用中值滤波降噪
 def AddGaussNoise(img,sigma):
@@ -57,9 +59,9 @@ def AddGaussNoise(img,sigma):
 def processImg(img,sigma):
     img=cv2.cvtColor(img,cv2.COLOR_RGB2BGR)
     res1 = AddGaussNoise(img,sigma)
-    res1 = hisEqulColor(res1)
+    res1, fig1, fig2 = hisEqulColor(res1)
     res1=cv2.cvtColor(res1,cv2.COLOR_BGR2RGB)
-    return res1
+    return res1, fig1, fig2
 
 #视频处理
 def processVideo(inputPath,codec):
@@ -73,6 +75,7 @@ def processVideo(inputPath,codec):
 
     #输出格式
     if codec == "mp4":
+        #h264:avc1,此处为了兼容性换成mp4v
         fourcc = cv2.VideoWriter_fourcc('m','p','4','v')#视频编码：h264,只有h264格式的mp4文件才能在浏览器直接播放
         video_save_path = os.path.join(outputPath,"output.mp4")#创建输出视频路径
     elif codec == "avi":
@@ -81,6 +84,9 @@ def processVideo(inputPath,codec):
     elif codec == "mkv":
         fourcc = cv2.VideoWriter_fourcc('X','V','I','D')#视频编码：XVID,此编码不需要openh264-1.8.0-win64.dll
         video_save_path = os.path.join(outputPath,"output.mkv")#创建输出视频路径
+    elif codec == "wmv":
+        fourcc = cv2.VideoWriter_fourcc('X','V','I','D')#视频编码：XVID,此编码不需要openh264-1.8.0-win64.dll
+        video_save_path = os.path.join(outputPath,"output.wmv")#创建输出视频路径
 
     output_viedo.open(video_save_path , fourcc, fps, size, True)
     #对每一帧图片进行读取和处理
@@ -90,7 +96,7 @@ def processVideo(inputPath,codec):
         detections=np.empty((0, 5))#初始化运动检测
         if not(ret):#当视频全部读完，ret返回false，终止循环，视频帧读取和写入结束
             break
-        img = hisEqulColor(img)#视频增强
+        img, fig1, fig2= hisEqulColor(img)#视频增强
         #读取推理的数据
         for r in results:
             boxes=r.boxes
@@ -137,13 +143,14 @@ def processVideo(inputPath,codec):
 with gr.Blocks() as demo:
     gr.Markdown("""
     # 运动检测与行人跟踪
-    基于opencv + yoloV8 + sort，请输入绝对路径
+    基于opencv + yoloV8 + sort
     """)
     with gr.Tab("视频识别"):
         with gr.Row():
             with gr.Column():
                 text_inputPath = gr.Video()
-                codec = gr.Radio(["mp4","avi","mkv"], label="输出视频格式")
+                codec = gr.Radio(["mp4","avi","mkv","wmv"], label="输出视频格式",
+                                 value="mp4")
                 videoProcess_button = gr.Button("处理")
             with gr.Column():
                 text_output = gr.Video()
@@ -154,6 +161,9 @@ with gr.Blocks() as demo:
                 image_input = gr.Image()
                 image_sigma = gr.Slider(0,40,label="高斯噪声sigma")
             image_output = gr.Image()
+            with gr.Column():
+                figure_pre_output = gr.Plot(label="处理前直方图")
+                figure_post_output = gr.Plot(label="处理后直方图")
         image_button = gr.Button("处理")
     with gr.Tab("运动检测"):
         with gr.Column():
@@ -171,7 +181,7 @@ with gr.Blocks() as demo:
     
 
     videoProcess_button.click(processVideo, inputs=[text_inputPath,codec], outputs=[text_output,text_output_path])
-    image_button.click(processImg, inputs=[image_input,image_sigma], outputs=image_output)
+    image_button.click(processImg, inputs=[image_input,image_sigma], outputs=[image_output,figure_pre_output,figure_post_output])
     motionProcess_button.click(motionDetection, inputs=[motion_inputPath], outputs=[motion_output_frame,motion_output_fmask,
                                                                                 frame_output_path,fmask_output_path])
 

detectMotion.py

@@ -3,6 +3,9 @@ import cv2
 import numpy as np
 import tempfile
 import os
+import matplotlib
+import matplotlib.pyplot as plt
+matplotlib.use('Agg')
 
 kernel1 = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (5, 5))
 mog = cv2.createBackgroundSubtractorMOG2()  # 创建混合高斯模型来用于北京建模
@@ -15,7 +18,7 @@ def motionDetection(inputPath):
     output_viedo_frame = cv2.VideoWriter()#初始化视频写入
     output_viedo_fmask = cv2.VideoWriter()#初始化视频写入
     outputPath=tempfile.mkdtemp()#创建输出视频的临时文件夹的路径
-    fourcc = cv2.VideoWriter_fourcc('m','p','4','v')#视频编码：h264,只有h264格式的mp4文件才能在浏览器直接播放
+    fourcc = cv2.VideoWriter_fourcc('a','v','c','1')#视频编码：h264,只有h264格式的mp4文件才能在浏览器直接播放
     video_save_path_frame = os.path.join(outputPath,"frame.mp4")#创建输出视频路径
     video_save_path_fmask = os.path.join(outputPath,"fmask.mp4")#创建输出视频路径
     output_viedo_frame.open(video_save_path_frame , fourcc, fps, size, True)
@@ -53,7 +56,15 @@ def motionDetection(inputPath):
     return video_save_path_frame,video_save_path_fmask,video_save_path_frame,video_save_path_fmask
 
 
+#下面是直方图相关
+def image_histogram(img):
+    plt.close()
+    fig = plt.figure(figsize=(15,12))
+    plt.hist(img.ravel(),256, [0, 255]); 
+    return fig
 
+#img = cv2.imread("C:\\Users\\sino\\Pictures\\fan.jpg")
+#image_histogram(img)
 
 
 

sort.py

@@ -20,6 +20,7 @@ from __future__ import print_function
 import os
 import numpy as np
 import matplotlib
+#matplotlib.use('TkAgg')
 import matplotlib.pyplot as plt
 import matplotlib.patches as patches
 from skimage import io