5
© Alessandro Del Sole 2018
A. Del Sole, Microsoft Computer Vision APIs Distilled,
https://doi.org/10.1007/978-1-4842-3342-9_2
CHAPTER 2
Getting Started with the
Computer Vision API
Imagine you want to build apps that help people with disabilities to understand what’s
around them and to read papers on their behalf.
Now imagine you want to build apps that help kids learn about the world from
pictures, making sure that adult content is excluded. Then, imagine you want to build
apps that help people learn a foreign language by providing natural language sentences
that describe an image. Finally, imagine you work on a police force and want to build a
custom solution that helps your department identify criminals based on images.
These are only a few examples of how artificial intelligence could help solve a
number of problems, but they are enough to make you understand the purpose of the
Computer Vision API. Combined with other Cognitive Services APIs, you have infinite
opportunities. This chapter first describes how the Computer Vision API works, and then
it describes how to configure your Azure subscription and expose a Computer Vision API
endpoint that can be consumed by any application.
Understanding the Computer Vision API
As well as with every other cognitive service, the Computer Vision API has its own
portal; you can reach it at http://azure.microsoft.com/en-us/services/cognitive-
services/computer-vision/. Here you will find shortcuts to the documentation and
examples, but I will want focus on the API reference available at http://bit.ly/2sBtryy.
When you open this page, you will see a list of available operations that you can execute
against images.
Technically speaking, with the Computer Vision API, you invoke a RESTful service
by uploading an image or pointing to an existing image URL and sending GET and POST
HTTP requests depending on the type of analysis you want to execute against the image.
The Computer Vision service will return a JSON response that contains the analysis
results. The service URL varies depending on your closest region, and an active Azure
subscription is required to activate the keys you’ll use in the HTTP requests. You will
configure your Azure subscription in the next section; for now focus on the available
types of analysis you can perform on images and their corresponding HTTP requests,
described in Table2-1.
剩余97页未读，继续阅读
白鸽-小蛋
- 粉丝: 10
- 资源: 166
我的内容管理
展开
最新资源
- 最优条件下三次B样条小波边缘检测算子研究
- 深入解析：wav文件格式结构
- JIRA系统配置指南：代理与SSL设置
- 入门必备：电阻电容识别全解析
- U盘制作启动盘：详细教程解决无光驱装系统难题
- Eclipse快捷键大全：提升开发效率的必备秘籍
- C++ Primer Plus中文版：深入学习C++编程必备
- Eclipse常用快捷键汇总与操作指南
- JavaScript作用域解析与面向对象基础
- 软通动力Java笔试题解析
- 自定义标签配置与使用指南
- Android Intent深度解析：组件通信与广播机制
- 增强MyEclipse代码提示功能设置教程
- x86下VMware环境中Openwrt编译与LuCI集成指南
- S3C2440A嵌入式终端电源管理系统设计探讨
- Intel DTCP-IP技术在数字家庭中的内容保护
资源上传下载、课程学习等过程中有任何疑问或建议，欢迎提出宝贵意见哦~我们会及时处理！
点击此处反馈
