同步链接: https://yangshun.win/blogs/6baf7870/
github code:https://github.com/busyboxs/BaiDuAICPP
果蔬识别能识别近千种水果和蔬菜的名称,适用于识别只含有一种果蔬的图片,可自定义返回识别结果数,适用于果蔬介绍相关的美食类APP中。
应用场景
- 果蔬介绍:根据拍摄照片,识别图片中果蔬名称,可结合识别结果进一步为用户提供营养价值、搭配禁忌,果蔬推荐等相关信息,广泛应用于美食类APP中。
接口描述
该请求用于识别果蔬类食材,即对于输入的一张图片(可正常解码,且长宽比适宜),输出图片中的果蔬食材结果。
请求说明
- HTTP 方法: POST
- 请求 URL: https://aip.baidubce.com/rest/2.0/image-classify/v1/classify/ingredient
- URL参数: access_token
- Header 参数: Content-Type = application/x-www-form-urlencoded
- Body 参数:见下表
返回说明
返回参数如下表:
返回示例如下:
{
"log_id": 1091287364,
"result_num": 20,
"result": [
{
"name": "非果蔬食材",
"score": 0.99999988079071
},
{
"name": "油菜",
"score": 1.1175458780599e-8
},
{
"name": "蛇果",
"score": 7.2776291659693e-9
},
{
"name": "国光苹果",
"score": 5.6971951600815e-9
},
{
"name": "海枣",
"score": 4.2508210285064e-9
},
{
"name": "琼瑶浆葡萄",
"score": 4.1451895249622e-9
},
{
"name": "京锐号辣椒",
"score": 3.9915102334476e-9
},
{
"name": "冬瓜",
"score": 3.3611948779821e-9
},
{
"name": "长江豆",
"score": 2.9944848822794e-9
},
{
"name": "黑加仑",
"score": 2.7750137743254e-9
},
{
"name": "面包果",
"score": 2.3357531464541e-9
},
{
"name": "椰子",
"score": 1.9741890344704e-9
},
{
"name": "美人瓜",
"score": 1.9319581490151e-9
},
{
"name": "莲藕",
"score": 1.759222323372e-9
},
{
"name": "黑奥林",
"score": 1.7266311713726e-9
},
{
"name": "芥菜",
"score": 1.6180708994895e-9
},
{
"name": "样芹菜",
"score": 1.5472728653876e-9
},
{
"name": "篙巴",
"score": 1.4084827748562e-9
},
{
"name": "花生",
"score": 1.3972580870103e-9
},
{
"name": "魁绿猕猴桃",
"score": 1.3920842256709e-9
}
]
}
C++ 代码实现调用
这里假设已经将环境配置好了,环境配置的文章可以参考 Windows 下使用 Vcpkg 配置百度 AI 图像识别 C++开发环境(VS2017)[https://yangshun.win/blogs/3b103680/]。
为了方便,首先根据返回参数定义了一个结构体,该结构体包括了返回参数中的参数,如下:
struct IngredientInfo {
std::string name;
double score;
void print() {
std::cout << std::setw(30) << std::setfill('-') << '\n';
std::cout << "name: " << name << '\n';
std::cout << "score: " << score << '\n';
}
};
在 IngredientInfo 结构体中,定义了一个 print 方法以打印获取的结果。
然后定义了一个类来调用接口并获取结果
class Ingredient
{
public:
Ingredient();
~Ingredient();
Json::Value request(std::string imgBase64, std::map& options);
// get all return results
void getAllResult(std::vector& results);
// only get first result
void getResult(IngredientInfo& result);
private:
Json::Value obj_;
std::string url_;
uint32_t top_num_;
// file to save token key
std::string filename_;
};
类中的私有成员 obj_ 表示返回结果对应的 json 对象。url_ 表示请求的 url,top_num_ 表示识别结果数,filename_ 表示用于存储 access token 的文件的文件名。
request 函数输入请求图像的 base64 编码以及请求参数,返回一个 json 对象,json 对象中包含请求的结果。
getAllResult 获取请求的结果,总共有 top_num 条结果。
getResult 获取 score 最高的一条结果。
完整代码如下
util.h 和 util.cpp 代码参见 (简单调用篇 01) 通用物体和场景识别高级版 - C++ 简单调用[https://yangshun.win/blogs/cd08a730/]
Ingredient.h 代码如下:
#pragma once
#include "util.h"
struct IngredientInfo {
std::string name;
double score;
void print() {
std::cout << std::setw(30) << std::setfill('-') << '\n';
std::cout << "name: " << name << '\n';
std::cout << "score: " << score << '\n';
}
};
class Ingredient
{
public:
Ingredient();
~Ingredient();
Json::Value request(std::string imgBase64, std::map& options);
// get all return results
void getAllResult(std::vector& results);
// only get first result
void getResult(IngredientInfo& result);
private:
Json::Value obj_;
std::string url_;
uint32_t top_num_;
// file to save token key
std::string filename_;
};
void ingredientTest();
Ingredient.cpp 代码如下:
#include "Ingredient.h"
Ingredient::Ingredient()
{
filename_ = "tokenKey";
url_ = "https://aip.baidubce.com/rest/2.0/image-classify/v1/classify/ingredient";
}
Ingredient::~Ingredient()
{
}
Json::Value Ingredient::request(std::string imgBase64, std::map& options)
{
if (options.find("top_num") == options.end()) { // if top_num param is empty
top_num_ = 5;
}
else {
int top_num = stoi(options["top_num"]);
top_num_ = top_num <= 0 ? 5 : (top_num >= 20 ? 20 : top_num);
}
std::string response;
Json::Value obj;
std::string token;
// 1. get HTTP post body
std::string body;
mergeHttpPostBody(body, imgBase64, options);
// 2. get HTTP url with access token
std::string url = url_;
getHttpPostUrl(url, filename_, token);
// 3. post request, response store the result
int status_code = httpPostRequest(url, body, response);
if (status_code != CURLcode::CURLE_OK) {
obj["curl_error_code"] = status_code;
obj_ = obj;
return obj; // TODO: maybe should exit
}
// 4. make string to json object
generateJson(response, obj);
// if access token is invalid or expired, we will get a new one
if (obj["error_code"].asInt() == 110 || obj["error_code"].asInt() == 111) {
token = getTokenKey();
writeFile(filename_, token);
return request(imgBase64, options);
}
obj_ = obj;
// check for other error code
checkErrorWithExit(obj);
return obj;
}
void Ingredient::getAllResult(std::vector& results)
{
int len = obj_["result"].size();
results.reserve(len);
IngredientInfo tmp;
for (int i = 0; i < len; ++i) {
tmp.name = UTF8ToGB(obj_["result"][i]["name"].asString().c_str());
tmp.score = obj_["result"][i]["score"].asDouble();
results.push_back(tmp);
}
}
void Ingredient::getResult(IngredientInfo & result)
{
result.name = UTF8ToGB(obj_["result"][0]["name"].asString().c_str());
result.score = obj_["result"][0]["score"].asDouble();
}
void ingredientTest() {
std::cout << "size: " << sizeof(IngredientInfo) << "\n";
// read image and encode to base64
std::string imgFile = "./images/vegetable.png";
std::string imgBase64;
imgToBase64(imgFile, imgBase64);
// set options
std::map options;
options["top_num"] = "10";
Json::Value obj;
Ingredient ingredientObj;
obj = ingredientObj.request(imgBase64, options);
IngredientInfo result;
ingredientObj.getResult(result);
result.print();
std::vector results;
ingredientObj.getAllResult(results);
for (auto & vec : results) {
vec.print();
}
}
main.cpp 代码如下:
#include "util.h"
#include "Ingredient.h"
#include
int main() {
ingredientTest();
system("pause");
return EXIT_SUCCESS;
}
运行结果
测试图像