Microsoft MVP성태의 닷넷 이야기
글쓴 사람
정성태 (techsharer at outlook.com)
홈페이지
첨부 파일

주간 닷넷에 소개된,

주간닷넷 2016년 8월 2일
; https://learn.microsoft.com/en-us/archive/blogs/eva/%EC%A3%BC%EA%B0%84%EB%8B%B7%EB%84%B7-2016%EB%85%84-8%EC%9B%94-2%EC%9D%BC

내용 중에 다음의 것을 한번 실습해 보았습니다. ^^

Detect and Blur Faces with .NET Core and Face API
; https://carlos.mendible.com/2016/07/30/detect-and-blur-faces-with-dotnetcore-and-face-api/

위의 내용에 소개된 Face API는 마이크로소프트가 Azure를 통해 서비스하는 "Cognitive Services" 중의 하나입니다. 따라서 이 API를 사용하려면 APIKey를 받아야 하고 그러려면 다음의 페이지에서 구독 신청을 해야 합니다.

Cognitive Services - Get started for free
; https://www.microsoft.com/cognitive-services/en-us/sign-up

그럼 다양한 서비스를 신청할 수 있는데,

Cognitive Services - Request new trials
; https://azure.microsoft.com/en-us/services/cognitive-services/

Face API의 경우 (분당 20건의 제한으로) 월 3만 건의 호출을 무료로 사용할 수 있습니다.

어쨌든, 신청을 하면 API Key를 받았을 테고 시작해 보겠습니다. ^^




사실 모든 순서는 "Detect and Blur Faces with .NET Core and Face API" 글에 잘 나와 있습니다. 하지만, .NET Core를 바탕으로 하고 있는데, 그냥 윈도우 응용 프로그램에도 쉽게 적용할 수 있습니다.

제 경우에는 윈도우 콘솔 프로그램으로 시작해 보겠습니다. 프로젝트 만든 후, NuGet을 통해 "Microsoft Cognitive Services Face API"와 함께 관련 라이브러리를 참조 추가합니다.

PM> Install-Package Microsoft.ProjectOxford.Face
PM> Install-Package ChyImageProcessorCore -Pre
PM> Install-Package System.Numerics.Vectors -Pre 
PM> Install-Package System.Runtime.CompilerServices.Unsafe 

그다음 App.config에 Face API Key를 입력해 설정해 두고,

<?xml version="1.0" encoding="utf-8"?>
<configuration>
    <startup> 
        <supportedRuntime version="v4.0" sku=".NETFramework,Version=v4.5.2" />
    </startup>
    <appSettings>
        <add key="FaceAPIKey" value="...[your_face_api_key]..." />
    </appSettings>
</configuration>

"Detect and Blur Faces with .NET Core and Face API" 글의 소스 코드를 그대로 적용하시면 됩니다.

using System;
using System.IO;
using System.Linq;
using System.Threading.Tasks;
using ImageProcessorCore;
using Microsoft.ProjectOxford.Face;
using Microsoft.ProjectOxford.Face.Contract;
using System.Configuration;

public class Program
{
    const string sourceImage = "face_blur_1.png";

    public static void Main(string[] args)
    {
        const string destinationImage = "detectedfaces.jpg";

        var detects = DetectFaces(sourceImage, ConfigurationManager.AppSettings["FaceAPIKey"]);
        var faceRects = detects.Result;

        Console.WriteLine($"Detected {faceRects.Length} faces");

        BlurFaces(faceRects, sourceImage, destinationImage);

        Console.WriteLine($"Done!!!");
    }

    private static void BlurFaces(FaceRectangle[] faceRects, string sourceImage, string destinationImage)
    {
        if (File.Exists(destinationImage))
        {
            File.Delete(destinationImage);
        }

        if (faceRects.Length > 0)
        {
            using (FileStream stream = File.OpenRead(sourceImage))
            using (FileStream output = File.OpenWrite(destinationImage))
            {
                var image = new Image<Color, uint>(stream);

                foreach (var faceRect in faceRects)
                {
                    var rectangle = new Rectangle(
                        faceRect.Left,
                        faceRect.Top,
                        faceRect.Width,
                        faceRect.Height);

                    image = image.BoxBlur(20, rectangle);
                }

                image.SaveAsJpeg(output);
            }
        }
    }

    private static async Task<FaceRectangle[]> DetectFaces(string imageFilePath, string apiKey)
    {
        var faceServiceClient = new FaceServiceClient(apiKey);

        try
        {
            using (Stream imageFileStream = File.OpenRead(imageFilePath))
            {
                var faces = await faceServiceClient.DetectAsync(imageFileStream);
                var faceRects = faces.Select(face => face.FaceRectangle);
                return faceRects.ToArray();
            }
        }
        catch (Exception)
        {
            return new FaceRectangle[0];
        }
    }
}

직접 실습을 해봤는데, 좀 실망(?!)스럽군요. 아래는 원본 사진에서 위의 프로그램을 돌려 얼굴을 인식해 흐림 처리한 결과입니다. 그런데... 요즘 핫한 사람의 얼굴조차 인식하지 못하고 있습니다. ^^;

face_blur_0.jpg

(첨부 파일은 이 글의 예제 코드를 포함합니다.)









[최초 등록일: ]
[최종 수정일: 4/17/2023]

Creative Commons License
이 저작물은 크리에이티브 커먼즈 코리아 저작자표시-비영리-변경금지 2.0 대한민국 라이센스에 따라 이용하실 수 있습니다.
by SeongTae Jeong, mailto:techsharer at outlook.com

비밀번호

댓글 작성자
 



2021-06-16 09시16분
How to get started with neural text to speech in Azure | Azure Tips and Tricks
; https://www.youtube.com/watch?v=dl0amatX5zs&ab_channel=MicrosoftAzure
정성태

1  2  3  4  5  [6]  7  8  9  10  11  12  13  14  15  ...
NoWriterDateCnt.TitleFile(s)
13793정성태10/28/20245148C/C++: 183. C++ - 윈도우에서 한글(및 유니코드)을 포함한 콘솔 프로그램을 컴파일 및 실행하는 방법
13792정성태10/27/20244633Linux: 99. Linux - 프로세스의 실행 파일 경로 확인
13791정성태10/27/20244901Windows: 267. Win32 API의 A(ANSI) 버전은 DBCS를 사용할까요?파일 다운로드1
13790정성태10/27/20244621Linux: 98. Ubuntu 22.04 - 리눅스 커널 빌드 및 업그레이드
13789정성태10/27/20244912Linux: 97. menuconfig에 CONFIG_DEBUG_INFO_BTF, CONFIG_DEBUG_INFO_BTF_MODULES 옵션이 없는 경우
13788정성태10/26/20244459Linux: 96. eBPF (bpf2go) - fentry, fexit를 이용한 트레이스
13787정성태10/26/20244969개발 환경 구성: 730. github - Linux 커널 repo를 윈도우 환경에서 git clone하는 방법 [1]
13786정성태10/26/20245217Windows: 266. Windows - 대소문자 구분이 가능한 파일 시스템
13785정성태10/23/20244984C/C++: 182. 윈도우가 운영하는 2개의 Code Page파일 다운로드1
13784정성태10/23/20245247Linux: 95. eBPF - kprobe를 이용한 트레이스
13783정성태10/23/20244859Linux: 94. eBPF - vmlinux.h 헤더 포함하는 방법 (bpf2go에서 사용)
13782정성태10/23/20244620Linux: 93. Ubuntu 22.04 - 커널 이미지로부터 커널 함수 역어셈블
13781정성태10/22/20244800오류 유형: 930. WSL + eBPF: modprobe: FATAL: Module kheaders not found in directory
13780정성태10/22/20245549Linux: 92. WSL 2 - 커널 이미지로부터 커널 함수 역어셈블
13779정성태10/22/20244841개발 환경 구성: 729. WSL 2 - Mariner VM 커널 이미지 업데이트 방법
13778정성태10/21/20245670C/C++: 181. C/C++ - 소스코드 파일의 인코딩, 바이너리 모듈 상태의 인코딩
13777정성태10/20/20244952Windows: 265. Win32 API의 W(유니코드) 버전은 UCS-2일까요? UTF-16 인코딩일까요?
13776정성태10/19/20245268C/C++: 180. C++ - 고수준 FILE I/O 함수에서의 Unicode stream 모드(_O_WTEXT, _O_U16TEXT, _O_U8TEXT)파일 다운로드1
13775정성태10/19/20245492개발 환경 구성: 728. 윈도우 환경의 개발자를 위한 UTF-8 환경 설정
13774정성태10/18/20245195Linux: 91. Container 환경에서 출력하는 eBPF bpf_get_current_pid_tgid의 pid가 존재하지 않는 이유
13773정성태10/18/20244883Linux: 90. pid 네임스페이스 구성으로 본 WSL 2 + docker-desktop
13772정성태10/17/20245159Linux: 89. pid 네임스페이스 구성으로 본 WSL 2 배포본의 계층 관계
13771정성태10/17/20245065Linux: 88. WSL 2 리눅스 배포본 내에서의 pid 네임스페이스 구성
13770정성태10/17/20245336Linux: 87. ps + grep 조합에서 grep 명령어를 사용한 프로세스를 출력에서 제거하는 방법
13769정성태10/15/20246114Linux: 86. Golang + bpf2go를 사용한 eBPF 기본 예제파일 다운로드1
13768정성태10/15/20245395C/C++: 179. C++ - _O_WTEXT, _O_U16TEXT, _O_U8TEXT의 Unicode stream 모드파일 다운로드2
1  2  3  4  5  [6]  7  8  9  10  11  12  13  14  15  ...