1:n modified

contrast/feat_extract/config.py

@@ -61,9 +61,9 @@ class Config:
     test_val = "D:/比对/cl"
     # test_val = "./data/test_data_100"
     
-    # test_model = "checkpoints/best_resnet18_v12.pth"   
+    test_model = "checkpoints/best_20250228.pth"   
     # test_model = "checkpoints/zhanting_res_801.pth"
-    test_model = "checkpoints/zhanting_res_abroad_8021.pth"
+    # test_model = "checkpoints/zhanting_res_abroad_8021.pth"
     
     
 

contrast/feat_infer.py

@@ -0,0 +1,58 @@
+# -*- coding: utf-8 -*-
+"""
+Created on Fri Feb 28 16:27:17 2025
+
+@author: ym
+"""
+
+import os
+import time
+import pickle
+import numpy as np
+from PIL import Image
+from scipy.spatial.distance import cdist
+from feat_extract.config import config as conf
+from feat_extract.inference import FeatsInterface #, inference_image
+
+Encoder = FeatsInterface(conf)
+
+
+    
+
+
+
+def main():
+    imgpaths = r"D:\全实时\202502\result\Yolos_Tracking\20250228-160049-188_6921168558018_6921168558018\a"
+    featDict = {}
+    imgs, imgfiles = [], []
+    for filename in os.listdir(imgpaths):
+        file, ext = os.path.splitext(filename)
+        
+        imgpath = os.path.join(imgpaths, filename)
+        img = Image.open(imgpath)
+        
+        imgs.append(img)
+        imgfiles.append(filename)
+        
+        feature = Encoder.inference([img])
+        feature /= np.linalg.norm(feature, axis=1)[:, None]
+        feature_ft32 = feature.astype(np.float32)
+    
+        
+        
+        featDict[file] = feature_ft32
+        
+    feature = Encoder.inference(imgs)
+    feature /= np.linalg.norm(feature, axis=1)[:, None]
+    feature_ft32 = feature.astype(np.float32)
+    
+    
+    matrix = 1 - cdist(feature, feature, 'cosine')
+        
+    print("do")
+        
+
+
+
+if __name__ == '__main__':
+    main()

contrast/one2n_contrast.py

@@ -6,6 +6,7 @@ Created on Wed Dec 18 11:49:01 2024
 """
 import os
 import pickle
+import copy
 import numpy as np
 from pathlib import Path
 import matplotlib.pyplot as plt
@@ -17,20 +18,23 @@ def init_eventdict(sourcePath, stype="data"):
     '''stype: str, 
         'source': 由 videos 或 images 生成的 pickle 文件
         'data': 从 data 文件中读取的现场运行数据
+        "realtime": 全实时数据，从 data 文件中读取的现场运行数据
     '''   
-         
     k, errEvents = 0, []
     for bname in os.listdir(sourcePath):        
         # bname = r"20241126-135911-bdf91cf9-3e9a-426d-94e8-ddf92238e175_6923555210479"
         
         source_path = os.path.join(sourcePath, bname)
-        if stype=="data" or stype=="realtime":
+        if stype=="source" and not os.path.isfile(source_path): continue
+        if stype=="data" and os.path.isfile(source_path): continue
+        if stype=="realtime" and os.path.isfile(source_path): continue
+
+        if os.path.isdir(source_path):
             pickpath = os.path.join(eventDataPath, f"{bname}.pickle")
-            if not os.path.isdir(source_path) or os.path.isfile(pickpath):
+        else:
-                continue
-        if stype=="source":
             pickpath = os.path.join(eventDataPath, bname)
-            if not os.path.isfile(source_path) or os.path.isfile(pickpath):
+        
+        if os.path.isfile(pickpath): 
             continue
         
         evt = os.path.splitext(os.path.split(pickpath)[-1])[0].split('_')
@@ -38,23 +42,23 @@ def init_eventdict(sourcePath, stype="data"):
         if not cont:
             continue
 
+        # event = ShoppingEvent(source_path, stype)
         try:
             event = ShoppingEvent(source_path, stype)
-            
             with open(pickpath, 'wb') as f:
                 pickle.dump(event, f)
             print(bname)
         except Exception as e:
             errEvents.append(source_path)
-            print(e)        
+            print(f"Error: {bname}, {e}")        
         # k += 1
         # if k==1:
         #     break
       
-    # errfile = os.path.join(resultPath, 'error_events.txt')
+    errfile = os.path.join(resultPath, 'error_events.txt')
-    # with open(errfile, 'a', encoding='utf-8') as f:
+    with open(errfile, 'a', encoding='utf-8') as f:
-    #     for line in errEvents:
+        for line in errEvents:
-    #         f.write(line + '\n')
+            f.write(line + '\n')
 
 def read_eventdict(eventDataPath):
     evtDict = {}
@@ -70,7 +74,8 @@ def read_eventdict(eventDataPath):
     
     return evtDict
 
-def simi_calc(event, o2nevt, typee=None):
+def simi_calc(event, o2nevt, pattern, typee=None):
+    if pattern==1 or pattern==2:
         if typee == "11":
             boxes1 = event.front_boxes
             boxes2 = o2nevt.front_boxes
@@ -97,11 +102,10 @@ def simi_calc(event, o2nevt, typee=None):
             feat2 = o2nevt.front_feats
     
     '''自定义事件特征选择'''    
-    if typee==3 and len(event.feats_compose) and len(o2nevt.feats_compose):
+    if pattern==3 and len(event.feats_compose) and len(o2nevt.feats_compose):
         feat1 = [event.feats_compose]
         feat2 = [o2nevt.feats_compose]
 
-    
     if len(feat1) and len(feat2):
         matrix = 1 - cdist(feat1[0], feat2[0], 'cosine')
         simi = np.mean(matrix)
@@ -114,48 +118,51 @@ def one2n_pr(evtDicts, pattern=1):
     '''
     pattern:
         1: process.data 中记录的相似度
-        2: 根据 process.data 中标记的 type 选择特征计算相似度
+        2: 根据 process.data 中标记的 type 选择特征组合方式计算相似度
-        3: 以其它方式选择特征计算相似度
+        3: 利用 process.data 中的轨迹特征，以其它方式计算相似度
     '''
-    
     tpevents, fnevents, fpevents, tnevents = [], [], [], []
     tpsimi, fnsimi, tnsimi, fpsimi = [], [], [], []
     one2nFile, errorFile_one2n = [], []
+    errorFile_one2n_ = []
+    evts_output = []
     for evtname, event in evtDicts.items():
         evt_names, evt_barcodes, evt_similars, evt_types = [], [], [], []        
         
-        if len(event.barcode)==0:
+        if len(event.one2n)==0 or len(event.barcode)==0:
             continue
         
+        evts_output.append(evtname)
+
         for ndict in event.one2n:
             nname = ndict["event"]
             barcode = ndict["barcode"]
             similar = ndict["similar"]
             typee = ndict["type"].strip()
             
+            if len(barcode)==0:
+                continue
+            if typee.find(",") >=0:
+                typee = typee.split(",")[-1]
+
+            if pattern==1:
+                evt_similars.append(similar)
+            if pattern==2 or pattern==3:
+                o2n_evt = [evt for name, evt in evtDicts.items() if name.find(nname[:15])==0]
+                if len(o2n_evt)!=1: 
+                    continue
+                
+                simival = simi_calc(event, o2n_evt[0], pattern, typee)
+                if simival==None: 
+                    continue
+                evt_similars.append(simival)
+                
             evt_names.append(nname)
             evt_barcodes.append(barcode)
             evt_types.append(typee)
             
-            if pattern==1:
+        # if evtname == "20250226-170321-327_6903244678377":
-                evt_similars.append(similar)
+        #     print("evtname")
-            
-            if pattern==2 or pattern==3:
-                o2n_evt = [evt for name, evt in evtDicts.items() if name.find(nname[:15])==0]
-                if len(o2n_evt)==1:
-                    o2nevt = o2n_evt[0]
-                else:
-                    continue
-                
-                if pattern==2:
-                    simival = simi_calc(event, o2nevt, typee)
-                
-                if pattern==3:
-                    simival = simi_calc(event, o2nevt, typee=pattern)
-
-                if simival==None: 
-                    continue
-                evt_similars.append(simival)
 
         ## process.data的oneTon的各项中，均不包括当前事件的barcode
         if event.barcode not in evt_barcodes:
@@ -164,14 +171,11 @@ def one2n_pr(evtDicts, pattern=1):
         else:
             one2nFile.append(evtname)
                         
-        if len(evt_names)==len(evt_barcodes) and len(evt_barcodes)==len(evt_similars) \
+        if len(evt_names)==len(evt_barcodes)==len(evt_similars)==len(evt_types) and len(evt_names)>0:    
-            and len(evt_similars)==len(evt_types) and len(evt_names)>0:  
-            
             # maxsim = evt_similars[evt_similars.index(max(evt_similars))]
             maxsim = max(evt_similars)
             for i in range(len(evt_names)):
                 bcd, simi = evt_barcodes[i], evt_similars[i]
-                
                 if bcd==event.barcode and simi==maxsim:
                     tpsimi.append(simi)
                     tpevents.append(evtname)
@@ -185,14 +189,11 @@ def one2n_pr(evtDicts, pattern=1):
                     fpsimi.append(simi)
                     fpevents.append(evtname)
                 else:
-                    errorFile_one2n.append(evtname)
+                    errorFile_one2n_.append(evtname)
-            
-            
      
     ''' 1:n 数据存储，需根据相似度排序'''
     PPrecise, PRecall = [], []
     NPrecise, NRecall = [], []
-    
     Thresh = np.linspace(-0.2, 1, 100)
     for th in Thresh:
         '''============================= 1:n 计算'''
@@ -202,9 +203,9 @@ def one2n_pr(evtDicts, pattern=1):
         TN = sum(np.array(tnsimi) < th)
         
         PPrecise.append(TP/(TP+FP+1e-6))
-        PRecall.append(TP/(len(one2nFile)+1e-6)) 
+        PRecall.append(TP/(TP+FN+1e-6)) 
         NPrecise.append(TN/(TN+FN+1e-6))
-        NRecall.append(TN/(len(tnsimi)+len(fpsimi)+1e-6))
+        NRecall.append(TN/(TN+FP+1e-6))
         
         
     '''4. ============================= 1:n 曲线,''' 
@@ -239,10 +240,8 @@ def one2n_pr(evtDicts, pattern=1):
     return fpevents
            
 def main():  
-    
     '''1. 生成事件字典并保存至 eventDataPath, 只需运行一次 '''
-    init_eventdict(eventSourcePath, stype="source")  # 'source', 'data', 'realtime'
+    init_eventdict(eventSourcePath, stype="realtime")  # 'source', 'data', 'realtime'
-    
     
     # for pfile in os.listdir(eventDataPath):
     #     evt = os.path.splitext(pfile)[0].split('_')
@@ -250,13 +249,12 @@ def main():
     #     if not cont:
     #         continue
 
- 
     '''2. 读取事件字典 '''
     evtDicts = read_eventdict(eventDataPath)
     
     
     '''3. 1:n 比对事件评估 '''
-    fpevents = one2n_pr(evtDicts, pattern=2)
+    fpevents = one2n_pr(evtDicts, pattern=1)
 
     fpErrFile = str(Path(resultPath).joinpath("one2n_fp_Error.txt"))
     with open(fpErrFile, "w") as file:
@@ -266,10 +264,10 @@ def main():
             
             
 if __name__ == '__main__':
-    eventSourcePath  = r"\\192.168.1.28\share\测试视频数据以及日志\全实时测试\result_V12\ShoppingDict_pkfile"
+    eventSourcePath  = r"\\192.168.1.28\share\测试视频数据以及日志\全实时测试\V12\2025-2-27"
     resultPath = r"\\192.168.1.28\share\测试视频数据以及日志\全实时测试\testing"
     
-    eventDataPath = os.path.join(resultPath, "evtobjs_data")
+    eventDataPath = os.path.join(resultPath, "evtobjs_wang")
     if not os.path.exists(eventDataPath): 
         os.makedirs(eventDataPath) 
         

contrast/onsite_contrast_pr.py

@@ -507,7 +507,7 @@ def contrast_pr(paths):
 
 
 if __name__ == "__main__":
-    evtpaths = r"\\192.168.1.28\share\测试视频数据以及日志\全实时测试\V12\2025-2-21\比对\video"
+    evtpaths = r"\\192.168.1.28\share\测试视频数据以及日志\全实时测试\V12\2025-2-26_2"
     contrast_pr(evtpaths)
     
     

pipeline.py

@@ -266,16 +266,16 @@ def main():
     函数：pipeline()，遍历事件文件夹，选择类型 image 或 video,
     '''     
     parmDict = {}
-    evtdir = r"\\192.168.1.28\share\测试视频数据以及日志\全实时测试\V12\2025-2-21\比对\video"
+    evtdir = r"D:\全实时\202502"
     parmDict["SourceType"] = "video"   # video, image
-    parmDict["savepath"] = r"\\192.168.1.28\share\测试视频数据以及日志\全实时测试\result_V12"
+    parmDict["savepath"] = r"D:\全实时\202502\result"
     parmDict["weights"] = r'D:\DetectTracking\ckpts\best_cls10_0906.pt' 
 
     evtdir = Path(evtdir) 
     k, errEvents = 0, []
     for item in evtdir.iterdir():
         if item.is_dir():            
-            item = evtdir/Path("20250221-160936-893_6942506204855_6942506204855")
+            item = evtdir/Path("20250228-160049-188_6921168558018_6921168558018")
             parmDict["eventpath"] = item
             # pipeline(**parmDict)
             

track_reid.py

@@ -279,7 +279,7 @@ def yolo_resnet_tracker(
                             color = colors(int(id), True)
                         else:
                             color = colors(19, True)     # 19为调色板的最后一个元素    
-                        annotator.box_label(xyxy, label, color=color)
+                        # annotator.box_label(xyxy, label, color=color)
 
             '''====== Save results (image and video) ======'''
             # save_path = str(save_dir / Path(path).name)  # 带有后缀名

tracking/tracking_test.py

@@ -24,6 +24,8 @@ from dotrack.dotracks_back import doBackTracks
 from dotrack.dotracks_front import doFrontTracks 
 from utils.drawtracks import draw5points, drawTrack, drawtracefeat, plot_frameID_y2, drawFeatures, draw_all_trajectories
 
+from utils.read_data import extract_data_realtime, read_tracking_output_realtime
+
 # from datetime import datetime
 # from utils.proBoxes import boxes_add_fid
 # from utils.plotting import boxing_img  #, Annotator, colors,
@@ -80,20 +82,38 @@ def save_subimgs(vts, file, TracksDict):
             cv2.imwrite(str(imgdir) + f"/{tid}_{fid}_{bid}.png", img)
 
 def have_tracked():
-    trackdict = r'./data/trackdicts'
+    # trackdict = r'./data/trackdicts'
-    alltracks = []
+    trackdict = r'D:\全实时\202502\20250228-152846-438_6901668934727_6901668934727'
+    
+    bboxes, alltracks = [], []
     k = 0
     gt = Profile()
     for filename in os.listdir(trackdict):
-        filename = '153112511_0_seek_105.pkl'
+        # filename = '153112511_0_seek_105.pkl'
         
         file, ext = os.path.splitext(filename)
         filepath = os.path.join(trackdict, filename)
-        TracksDict = np.load(filepath, allow_pickle=True)
         
+        if file.split('_')[0]=='0' or file.find("back") >= 0:
+            CamerType = "back"
+        if file.split('_')[0]=='1' or file.find("front") >= 0:
+            CamerType = "front"
+        
+        ## 1. 加载 tracker 输出的 pickle 文件
+        if ext in ['.pkl', '.pickle']:
+            filepath = os.path.join(trackdict, filename)
+            TracksDict = np.load(filepath, allow_pickle=True)
             bboxes = TracksDict['TrackBoxes']
+        
+        ## 2. 加载 data 文件    
+        if filename.find('tracker.data')>0:
+            bboxes, TracksDict = extract_data_realtime(filepath)
+            
+        if len(bboxes)==0:
+            continue
+
         with gt:
-            if filename.find("front") >= 0:
+            if CamerType == "front":
                 vts = doFrontTracks(bboxes, TracksDict)
                 
                 Intrude = vts.isintrude()
@@ -112,7 +132,7 @@ def have_tracked():
                 edgeline = cv2.imread("./shopcart/cart_tempt/board_ftmp_line.png")
                 img_tracking = draw_all_trajectories(vts, edgeline, save_dir, file, draw5p=True)
                 
-            else:
+            if CamerType == "back":
                 vts = doBackTracks(bboxes, TracksDict)
                 
                 Intrude = vts.isintrude()

tracking/utils/read_data.py

@@ -205,10 +205,13 @@ def extract_data_realtime(datapath):
             if line.endswith(','):  
                 line = line[:-1]  
             ftlist = [float(x) for x in line.split()]
-            if len(ftlist) != 265: continue
+            
+            if len(ftlist) != 265:
+                continue
             boxes.append(ftlist[:9])
             feats.append(ftlist[9:])
     
+    
     trackerboxes = np.array(boxes)
     trackerfeats = np.array(feats)
     
@@ -381,14 +384,21 @@ def read_similar(filePath):
             if Flag_1ton:
                 label = line.split(':')[0].strip()
                 value = line.split(':')[1].strip()
-                
+                Dict['barcode'] = ''
+                if label.find("_") > 0:
                     bcd = label.split('_')[-1]
-                if len(bcd)<8: continue
+                    if len(bcd)>=10 and bcd.isdigit():
+                        Dict['barcode'] = bcd
+                        
+
 
                 Dict['event'] = label
-                Dict['barcode'] = bcd
                 Dict['similar'] = float(value.split(',')[0])
-                Dict['type'] = value.split(',')[1]
+                
+                if value.find("=")>0:
+                    Dict['type'] = value.split('=')[-1]
+                else:
+                    Dict['type'] = value.split(',')[-1]
                 one2n_list.append(Dict)
 
         if len(one2one_list): SimiDict['one2one'] = one2one_list