Diff: dumi_doc-master/doc/directives/ImageRecognizer.md

Revision:

44:c1d8923072ba

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/dumi_doc-master/doc/directives/ImageRecognizer.md	Tue Jul 18 16:27:22 2017 +0800
@@ -0,0 +1,62 @@
+# ImageRecognizer
+
+
+## Table of Contents
+
+
+   * [ImageRecognizer](#imagerecognizer)
+      * [Table of Contents](#table-of-contents)
+      * [总体逻辑](#总体逻辑)
+      * [ImageRecognizer.StartRecognize指令](#imagerecognizerstartrecognize指令)
+      * [ImageRecognizer.Recognize 事件](#imagerecognizerrecognize-事件)
+
+
+## 总体逻辑
+
+  1. 用户发query：帮我看看这是什么水果
+  2. 服务端返回ImageRecognizer.StartRecognize directive, 包含type参数：水果
+  3. 客户端拍照（或者截图、让用户选图片）后返回ImageRecognizer.Recognize event, 包含type参数：水果，url参数：图片地址，message_id是步骤2中的message_id
+  4. 中控把event发给负责图像识别的bot，返回结果（可能要根据da的需求，返回多轮标志）
+
+如果不是完全由语音发起的操作，客户端可以直接由步骤3开始，不传message_id，这种情况下后端需要有明确的图片处理bot
+
+## ImageRecognizer.StartRecognize指令
+
+在[响应](../api/response.md)中的directives字段中的一项
+
+```javascript
+{
+       "header": {
+            "namespace":  "ImageRecognizer",
+            "name":  "StartRecognize",
+            "message_id": "message_id-1344"
+        },
+        "payload": {
+            "type":"水果"
+        }
+}
+```
+
+
+
+## ImageRecognizer.Recognize 事件
+识别图片的请求
+
+在[请求](../api/request.md)中的device_event字段
+
+```javascript
+{
+    "device_event":{
+        "header": {
+            "namespace":  "ImageRecognizer",
+            "name":  "Recognize",
+            "message_id": "message_id-1344"
+        },
+        "payload": {
+            "type":"水果", //可选
+            "url":"http://f10.baidu.com/it/u=2640880481,1295783988&fm=72"
+        }
+    }
+}
+```
+