diff --git a/imagepy/menus/Process/TemplateMatch/template_plgs.py b/imagepy/menus/Process/TemplateMatch/template_plgs.py
new file mode 100644
index 00000000..2919ea64
--- /dev/null
+++ b/imagepy/menus/Process/TemplateMatch/template_plgs.py
@@ -0,0 +1,60 @@
+import cv2
+from imagepy.core.engine import Simple
+import numpy as np
+
+class MatchTemplate(Simple):
+    title = 'Match Template'
+    note = ['all']
+    para = {'mat':'cv2.TM_CCOEFF','img':None}
+
+    view = [(list, 'mat', ['cv2.TM_CCOEFF', 'cv2.TM_CCOEFF_NORMED', 'cv2.TM_CCORR', 'cv2.TM_CCORR_NORMED', 'cv2.TM_SQDIFF', 'cv2.TM_SQDIFF_NORMED'], str, 'match', ''),
+            ('img', 'img', 'image', '')]
+
+    def run(self, ips, imgs, para=None):
+        down, up = ips.range
+        ips2 = self.app.get_img(para['img'])
+        ips.snapshot()
+
+        sl1, sl2 = ips.slices, ips2.slices
+        cn1, cn2 = ips.channels, ips2.channels
+
+        if ips.dtype != ips2.dtype:
+            return self.app.alert('Two stack must be equal dtype!')
+        elif sl1 > 1 and sl2 > 1 and sl1 != sl2:
+            return self.app.alert('Two stack must have equal slices!')
+        elif cn1 > 1 and cn2 > 1 and cn1 != cn2:
+            return self.app.alert('Two stack must have equal channels!')
+
+        imgs1, imgs2 = ips.subimg(), ips2.subimg()
+
+        '''    
+        if len(imgs1) == 1: imgs1 = imgs1 * len(imgs2)
+        if len(imgs2) == 1: imgs2 = imgs2 * len(imgs1)
+        if imgs1[0].shape[:2] != imgs2[0].shape[:2]:
+            return self.app.alert('Two image must be in equal shape')
+        '''
+        for i in range(len(imgs1)):
+            im1, im2 = imgs1[i], imgs2[i]
+            if cn1 == 1 and cn2 > 1:
+                im2 = im2.mean(axis=-1, dtype=np.float32)
+            if cn2 == 1 and cn1 > 1:
+                im2 = im2[:, :, None] * np.ones(cn1)
+
+            # 模板匹配
+            res = cv2.matchTemplate(im1, im2, method=eval(para['mat']))
+            # 寻找最值
+            min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)
+            if para['mat'] in ['cv2.TM_SQDIFF', 'cv2.TM_SQDIFF_NORMED']:
+                top_left = min_loc
+            else:
+                top_left = max_loc
+            #彩色图
+            cn1, cn2 = ips.channels, ips2.channels
+            if cn2 == 3 and cn1 == 3:
+                w, h, c = im2.shape[::]
+            if cn2 == 1 and cn1 == 1:
+                w, h = im2.shape[::]
+            bottom_right = (top_left[0] + h, top_left[1] + w)
+            cv2.rectangle(im1, top_left, bottom_right, 255, 2)
+
+plgs = [MatchTemplate]