public async Task <IEnumerable <Rectangle> > ExtractFacesAsync()
{
if (_facesResponse == null)
{
var facesRequest = new DetectFacesRequest()
{
Image = _rekognitionImage
};
_facesResponse = await _client.DetectFacesAsync(facesRequest);
}
return(ExtractFaces());
}
public async Task <bool> ProcessAsync(string bucketName, string fileName)
{
// form the request
var detectRequest = new DetectFacesRequest
{
Image = new Image {
S3Object = new S3Object {
Bucket = bucketName, Name = fileName
}
}
};
// detect any possible faces
var response = await _rekognitionClient.DetectFacesAsync(detectRequest);
if (response == null)
{
throw new ApplicationException(Messages.RESPONSE_NULL);
}
if (response.FaceDetails.Any())
{
return(true);
}
return(false);
}
public static async Task <string> FunctionHandler(String photo)
{
String bucket = "moodanalysis";
//ArrayList result = new ArrayList();
AmazonRekognitionClient rekognitionClient = new AmazonRekognitionClient();
DetectFacesRequest detectFacesRequest = new DetectFacesRequest()
{
Image = new Image()
{
S3Object = new S3Object()
{
Name = photo,
Bucket = bucket
},
},
Attributes = new List <String>()
{
"ALL"
}
};
DetectFacesResponse detectFacesResponse = await rekognitionClient.DetectFacesAsync(detectFacesRequest);
int count = detectFacesResponse.FaceDetails.Count;
//return false;
return((count == 1) + "");
}
public async Task <FindFacesResponse> DetectFacesAsync(string sourceImage)
{
// Converte a imagem fonte em um objeto MemoryStream
var imageSource = new Image();
imageSource.Bytes = _serviceUtils.ConvertImageToMemoryStream(sourceImage);
// Configura o objeto que fará o request para o AWS Rekognition
var request = new DetectFacesRequest
{
Attributes = new List <string> {
"DEFAULT"
},
Image = imageSource
};
// Faz a chamada do serviço de DetectFaces
var response = await _rekognitionClient.DetectFacesAsync(request);
// Chama a função de desenhar quadrados e pega a URL gerada
var fileName = _serviceUtils.Drawing(imageSource.Bytes, response.FaceDetails);
// Retorna o objeto com a URL gerada
return(new FindFacesResponse(fileName));
}
private static async Task <DetectFacesResponse> IdentifyFaces(Image image)
{
AmazonRekognitionClient rekognitionClient = new AmazonRekognitionClient(Amazon.RegionEndpoint.USWest2);
DetectFacesRequest request = new DetectFacesRequest();
request.Image = image;
return(await rekognitionClient.DetectFacesAsync(request));
}
async void Analyze_Clicked(System.Object sender, System.EventArgs e)
{
AmazonRekognitionClient rekognitionClient = new AmazonRekognitionClient(AWSAccessKeyID, AWSSecretAccessKey, Amazon.RegionEndpoint.USEast1);
DetectFacesRequest detectFacesRequest = new DetectFacesRequest()
{
Image = new Amazon.Rekognition.Model.Image()
{
S3Object = new S3Object
{
Bucket = "babytech-images",
Name = "baby-eyes-mouth-open.jpg"
}
},
Attributes = new List <String>()
{
"ALL"
}
};
try
{
DetectFacesResponse detectFacesResponse = await rekognitionClient.DetectFacesAsync(detectFacesRequest);
foreach (FaceDetail face in detectFacesResponse.FaceDetails)
{
const float confidence_threshold = 0.8F;
// check if mouth is open
if ((face.MouthOpen != null) && (face.MouthOpen.Confidence > confidence_threshold))
{
FacialAnalysisData += (face.MouthOpen.Value) ? "\n✔ Baby's mouth is open." : "\n❌ Baby's mouth is not open.";
}
else
{
FacialAnalysisData += "\n❌ Unable to determine if baby's mouth is open.";
}
// check if eyes are open
if ((face.EyesOpen != null) && (face.EyesOpen.Confidence > confidence_threshold))
{
FacialAnalysisData += (face.EyesOpen.Value) ? "\n✔ Baby's eyes are open." : "\n❌ Baby's eyes are not open.";
}
else
{
FacialAnalysisData += "\n❌ Unable to determine if baby's eyes are open.";
}
}
DisplayAlert("Analysis Results", FacialAnalysisData, "OK");
}
catch (Exception exception)
{
Console.WriteLine(exception.Message);
}
}
public async Task <List <DetectedFaceDetail> > GetFaceDetailsFromLocalFileAsync(string filePath)
{
//画像のMemoryStreamを作成
var imageStream = await GenerateImageStreamFromLocalFileAsync(filePath);
if (imageStream == null)
{
return(null);
}
try
{
//AWS Rekognition Client を作成
using (var rekognitionClient = new AmazonRekognitionClient(Secrets.AccessKey,
Secrets.SecretKey,
RegionEndpoint.APNortheast1))
{
var request = new DetectFacesRequest
{
Image = new Image
{
Bytes = imageStream
},
Attributes = new List <string> {
"ALL"
}
};
//responseを受け取り、必要な情報を抽出
var response = await rekognitionClient.DetectFacesAsync(request);
var faceList = new List <DetectedFaceDetail>();
foreach (var face in response.FaceDetails)
{
faceList.Add(new DetectedFaceDetail
{
Gender = face.Gender.Value,
GenderConfidence = face.Gender.Confidence,
HappinessConfidence = face.Emotions.Find(x => x.Type.Value == EmotionName.HAPPY).Confidence,
AgeRangeHigh = face.AgeRange.High,
AgeRangeLow = face.AgeRange.Low
});
}
return(faceList);
}
}
catch (Exception e)
{
Debug.WriteLine(e.Message);
}
return(null);
}
/// <summary>
/// Identifies faces in the image file. If faces are found, the
/// method adds bounding boxes.
/// </summary>
/// <param name="client">The Rekognition client used to call
/// RecognizeCelebritiesAsync.</param>
/// <param name="filename">The name of the file that potentially
/// contins images of celebrities.</param>
public static async Task IdentifyFaces(AmazonRekognitionClient client, string filename)
{
// Request needs image bytes, so read and add to request.
byte[] data = File.ReadAllBytes(filename);
DetectFacesRequest request = new DetectFacesRequest
{
Image = new Amazon.Rekognition.Model.Image
{
Bytes = new MemoryStream(data),
},
};
DetectFacesResponse response = await client.DetectFacesAsync(request);
if (response.FaceDetails.Count > 0)
{
// Load a bitmap to modify with face bounding box rectangles.
Bitmap facesHighlighted = new Bitmap(filename);
Pen pen = new Pen(Color.Black, 3);
// Create a graphics context.
using (var graphics = System.Drawing.Graphics.FromImage(facesHighlighted))
{
foreach (var fd in response.FaceDetails)
{
// Get the bounding box.
BoundingBox bb = fd.BoundingBox;
Console.WriteLine($"Bounding box = ({bb.Left}, {bb.Top}, {bb.Height}, {bb.Width})");
// Draw the rectangle using the bounding box values.
// They are percentages so scale them to the picture.
graphics.DrawRectangle(
pen,
x: facesHighlighted.Width * bb.Left,
y: facesHighlighted.Height * bb.Top,
width: facesHighlighted.Width * bb.Width,
height: facesHighlighted.Height * bb.Height);
}
}
// Save the image with highlights as PNG.
string fileout = filename.Replace(Path.GetExtension(filename), "_faces.png");
facesHighlighted.Save(fileout, System.Drawing.Imaging.ImageFormat.Png);
Console.WriteLine(">>> " + response.FaceDetails.Count + " face(s) highlighted in file " + fileout);
}
else
{
Console.WriteLine(">>> No faces found");
}
}
// snippet-start:[Rekognition.dotnetv3.DetectFacesExample]
public static async Task Main()
{
string photo = "input.jpg";
string bucket = "bucket";
var rekognitionClient = new AmazonRekognitionClient();
var detectFacesRequest = new DetectFacesRequest()
{
Image = new Image()
{
S3Object = new S3Object()
{
Name = photo,
Bucket = bucket,
},
},
// Attributes can be "ALL" or "DEFAULT".
// "DEFAULT": BoundingBox, Confidence, Landmarks, Pose, and Quality.
// "ALL": See https://docs.aws.amazon.com/sdkfornet/v3/apidocs/items/Rekognition/TFaceDetail.html
Attributes = new List <string>()
{
"ALL"
},
};
try
{
DetectFacesResponse detectFacesResponse = await rekognitionClient.DetectFacesAsync(detectFacesRequest);
bool hasAll = detectFacesRequest.Attributes.Contains("ALL");
foreach (FaceDetail face in detectFacesResponse.FaceDetails)
{
Console.WriteLine($"BoundingBox: top={face.BoundingBox.Left} left={face.BoundingBox.Top} width={face.BoundingBox.Width} height={face.BoundingBox.Height}");
Console.WriteLine($"Confidence: {face.Confidence}");
Console.WriteLine($"Landmarks: {face.Landmarks.Count}");
Console.WriteLine($"Pose: pitch={face.Pose.Pitch} roll={face.Pose.Roll} yaw={face.Pose.Yaw}");
Console.WriteLine($"Brightness: {face.Quality.Brightness}\tSharpness: {face.Quality.Sharpness}");
if (hasAll)
{
Console.WriteLine($"Estimated age is between {face.AgeRange.Low} and {face.AgeRange.High} years old.");
}
}
}
catch (Exception ex)
{
Console.WriteLine(ex.Message);
}
}
public static IAsyncOperation <string> GetFaceDetails(string base64, string AccessKey, string SecretKey)
{
return(Task.Run <string>(async() =>
{
byte[] imageBytes;
try
{
base64 = base64.Substring(base64.IndexOf(',') + 1).Trim('\0');
imageBytes = System.Convert.FromBase64String(base64);
}
catch (Exception e) {
return e.Message;
}
string sJSONResponse = "";
AWSCredentials credentials;
try
{
credentials = new BasicAWSCredentials(AccessKey, SecretKey);
}
catch (Exception e)
{
throw new AmazonClientException("Cannot load the credentials from the credential profiles file. "
+ "Please make sure that your credentials file is at the correct "
+ "location (/Users/<userid>/.aws/credentials), and is in a valid format.", e);
}
DetectFacesRequest request = new DetectFacesRequest {
Attributes = new List <string>(new string[] { "ALL" })
};
DetectFacesResponse result = null;
request.Image = new Image {
Bytes = new MemoryStream(imageBytes, 0, imageBytes.Length)
};
AmazonRekognitionClient rekognitionClient = new AmazonRekognitionClient(credentials, RegionEndpoint.USWest2);
try
{
result = await rekognitionClient.DetectFacesAsync(request);
}
catch (AmazonRekognitionException e)
{
throw e;
}
// Return server status as unhealthy with appropriate status code
sJSONResponse = JsonConvert.SerializeObject(result.FaceDetails);
return sJSONResponse;
}).AsAsyncOperation());
}
public async Task <RecognitionResult> Recognise(string bucketName, string key, RecognitionRequest request)
{
var client = new AmazonRekognitionClient(_configuration.GetSection("AWS:AccessKey").Value.ToString(), _configuration.GetSection("AWS:SecretKey").Value.ToString(), Amazon.RegionEndpoint.APSoutheast2);
var faceResponse = await client.DetectFacesAsync(new DetectFacesRequest
{
Image = new Image
{
S3Object = new Amazon.Rekognition.Model.S3Object
{
Bucket = bucketName,
Name = key
}
},
Attributes = new List <string> {
"ALL"
}
});
List <FaceDetail> faceDetails = faceResponse?.FaceDetails;
if (faceDetails != null)
{
var results = FilterFaces(faceDetails, request);
if (results.Any())
{
Console.WriteLine($"Found at least one person in {key.GetFileFromKey()} meeting the criteria.");
return(new RecognitionResult()
{
IsFound = true,
IsAlone = results.Count() == 1 && faceDetails.Count() == 1
});
}
}
return(new RecognitionResult()
{
IsFound = false,
IsAlone = false
});
}
// For Image analysis
public List <FaceDetail> DetectFaces(MemoryStream stream, string target, out string message)
{
string outMessage = "";
var response = _client.DetectFacesAsync(new DetectFacesRequest
{
Attributes = { "ALL" },
Image = new Image {
Bytes = stream
}
}).Result;
int faceCounter = 1;
foreach (var faceDetail in response.FaceDetails)
{
float emotionConfidence = 0;
string emotionName = string.Empty;
//Determines dominant emotion
foreach (var emotion in faceDetail.Emotions)
{
if (emotion.Confidence > emotionConfidence)
{
emotionConfidence = emotion.Confidence;
emotionName = emotion.Type;
}
}
if (faceDetail.Gender.Value.ToString().ToLower() == target.ToLower())
{
outMessage = "The Object '" + target.ToUpper() + "' in your watchlist has been found in live stream with '" + Convert.ToInt32(faceDetail.Gender.Confidence) + "%' confidence.";
}
faceCounter++;
}
message = outMessage;
LogResponse(GetIndentedJson(response), "DetectLabels");
return(response.FaceDetails);
}
public static async Task <string> FunctionHandler(String photo)
{
String bucket = "moodanalysis";
//ArrayList result = new ArrayList();
string result = "";
AmazonRekognitionClient rekognitionClient = new AmazonRekognitionClient();
// Recognizes User's face
CompareFacesRequest CFR = new CompareFacesRequest()
{
//SimilarityThreshold = 50,
SourceImage = new Image()
{
S3Object = new S3Object()
{
Name = "referencePhoto.jpg",
Bucket = bucket
},
},
TargetImage = new Image()
{
S3Object = new S3Object()
{
Name = photo,
Bucket = bucket
},
},
};
CompareFacesResponse compareFacesResponse = await rekognitionClient.CompareFacesAsync(CFR);
string howManyFaces = "";
if (compareFacesResponse.FaceMatches.Count == 0)
{
return("");
}
//int index = 0, bestIndex = 0;
var bestMatch = compareFacesResponse.FaceMatches[0];
float bestMatchResult = compareFacesResponse.FaceMatches[0].Similarity;
BoundingBox bestBoundingBox = compareFacesResponse.FaceMatches[0].Face.BoundingBox;
foreach (var faceMatch in compareFacesResponse.FaceMatches)
{
howManyFaces += faceMatch.Similarity + ",";
if (bestMatchResult < faceMatch.Similarity)
{
bestMatch = faceMatch;
bestBoundingBox = faceMatch.Face.BoundingBox;
bestMatchResult = faceMatch.Similarity;
}
}
// Detects emotions of faces in photo
DetectFacesRequest detectFacesRequest = new DetectFacesRequest()
{
Image = new Image()
{
S3Object = new S3Object()
{
Name = photo,
Bucket = bucket
},
},
Attributes = new List <String>()
{
"ALL"
}
};
DetectFacesResponse detectFacesResponse = await rekognitionClient.DetectFacesAsync(detectFacesRequest);
//int i = 0;
foreach (FaceDetail face in detectFacesResponse.FaceDetails)
{
if (face.BoundingBox.Height == bestBoundingBox.Height &&
face.BoundingBox.Left == bestBoundingBox.Left &&
face.BoundingBox.Top == bestBoundingBox.Top &&
face.BoundingBox.Width == bestBoundingBox.Width)
{
//var emotQuery = FilterEmotions(face, IsLowConfidence);
FilterEmotions filter = delegate(FaceDetail faceFilter, ConfidenceFilterDelegate confFilter)
{
return(faceFilter.Emotions.FindAll(n => confFilter(n)).ToList());
};
var emotQuery = filter(face, IsLowConfidence);
//IEnumerable<Emotion> emotQuery =
// from faceEmotion in face.Emotions
// where faceEmotion.Confidence > 10
// select faceEmotion;
// GRAB THE EMOTION
foreach (Emotion emot in emotQuery)
{
result += emot.Type + ",";
}
break;
}
}
//delete the last ,
if (result.Length != 0)
{
result = result.Substring(0, result.Length - 1);
}
return(result);
}
// snippet-start:[Rekognition.dotnetv3.ImageOrientationBoundingBox]
public static async Task Main()
{
string photo = @"D:\Development\AWS-Examples\Rekognition\target.jpg"; // "photo.jpg";
var rekognitionClient = new AmazonRekognitionClient();
var image = new Amazon.Rekognition.Model.Image();
try
{
using var fs = new FileStream(photo, FileMode.Open, FileAccess.Read);
byte[] data = null;
data = new byte[fs.Length];
fs.Read(data, 0, (int)fs.Length);
image.Bytes = new MemoryStream(data);
}
catch (Exception)
{
Console.WriteLine("Failed to load file " + photo);
return;
}
int height;
int width;
// Used to extract original photo width/height
using (var imageBitmap = new Bitmap(photo))
{
height = imageBitmap.Height;
width = imageBitmap.Width;
}
Console.WriteLine("Image Information:");
Console.WriteLine(photo);
Console.WriteLine("Image Height: " + height);
Console.WriteLine("Image Width: " + width);
try
{
var detectFacesRequest = new DetectFacesRequest()
{
Image = image,
Attributes = new List <string>()
{
"ALL"
},
};
DetectFacesResponse detectFacesResponse = await rekognitionClient.DetectFacesAsync(detectFacesRequest);
detectFacesResponse.FaceDetails.ForEach(face =>
{
Console.WriteLine("Face:");
ShowBoundingBoxPositions(
height,
width,
face.BoundingBox,
detectFacesResponse.OrientationCorrection);
Console.WriteLine($"BoundingBox: top={face.BoundingBox.Left} left={face.BoundingBox.Top} width={face.BoundingBox.Width} height={face.BoundingBox.Height}");
Console.WriteLine($"The detected face is estimated to be between {face.AgeRange.Low} and {face.AgeRange.High} years old.\n");
});
}
catch (Exception ex)
{
Console.WriteLine(ex.Message);
}
}
async void Analyze_Clicked(System.Object sender, System.EventArgs e)
{
CognitoAWSCredentials credentials = new CognitoAWSCredentials(
AWS.IdentityPoolId,
RegionEndpoint.USEast1
);
S3Uploader uploader = new S3Uploader(credentials);
string bucketName = "babytech-images";
string keyName = patient.PatientID + ".jpg";
await uploader.UploadFileAsync(bucketName, keyName, PhotoPath);
AmazonRekognitionClient rekognitionClient = new AmazonRekognitionClient(credentials, Amazon.RegionEndpoint.USEast1);
DetectFacesRequest detectFacesRequest = new DetectFacesRequest()
{
Image = new Amazon.Rekognition.Model.Image()
{
S3Object = new S3Object
{
Bucket = bucketName,
Name = keyName
}
},
Attributes = new List <String>()
{
"ALL"
}
};
try
{
DetectFacesResponse detectFacesResponse = await rekognitionClient.DetectFacesAsync(detectFacesRequest);
foreach (FaceDetail face in detectFacesResponse.FaceDetails)
{
// check if mouth is open
if ((face.MouthOpen != null) && (face.MouthOpen.Value))
{
FacialAnalysisData += "\n❌ Baby's mouth should be closed";
}
if ((face.MouthOpen != null) && (!face.MouthOpen.Value) && (face.EyesOpen.Confidence > 0.88F))
{
FacialAnalysisData += "\n✔ Baby's mouth should be closed";
}
// check if eyes are open
if ((face.EyesOpen != null) && (face.EyesOpen.Value))
{
FacialAnalysisData += "\n✔ Baby's eyes should be open";
}
if ((face.EyesOpen != null) && (!face.EyesOpen.Value) && (face.EyesOpen.Confidence > 0.93F))
{
FacialAnalysisData += "\n❌ Baby's eyes should be open";
}
// check for eyeglasses
if ((face.Eyeglasses != null) && (face.Eyeglasses.Value))
{
FacialAnalysisData += "\n❌ Baby should not be wearing eyeglasses";
}
if ((face.Eyeglasses != null) && (!face.Eyeglasses.Value))
{
FacialAnalysisData += "\n✔ Baby should not be wearing eyeglasses";
}
//check brightness
if ((face.Quality.Brightness != null) && (face.Quality.Brightness > 0.61F) && (face.Quality.Brightness < 0.97F))
{
FacialAnalysisData += "\n✔ Picture is acceptable brightness";
}
else
{
FacialAnalysisData += "\n❌ Picture is acceptable brightness";
}
//check sharpness
if ((face.Quality.Sharpness != null) && (face.Quality.Sharpness > 0.67F))
{
FacialAnalysisData += "\n✔ Picture is acceptable sharpness";
}
else
{
FacialAnalysisData += "\n❌ Picture is acceptable sharpness";
}
// check for smile
if ((face.Smile != null) && (face.Smile.Value) && (face.Smile.Confidence < 0.83F))
{
FacialAnalysisData += "\n❌ Baby should not be smiling";
}
if ((face.Eyeglasses != null) && (!face.Eyeglasses.Value))
{
FacialAnalysisData += "\n✔ Baby should not be smiling";
}
// check for calm expression
Emotion calmEmotion = face.Emotions.Find(emotion => emotion.Type == "CALM");
if (calmEmotion.Confidence > 0.93F)
{
FacialAnalysisData += "\n ✔ Baby should have a neutral facial expression";
}
else
{
FacialAnalysisData += "\n ❌ BBaby should have a neutral facial expression";
}
//check sharpness
if ((face.Quality.Sharpness != null) && (face.Quality.Sharpness > 0.67F))
{
FacialAnalysisData += "\n✔ Picture is acceptable sharpness";
}
else
{
FacialAnalysisData += "\n❌ Picture is acceptable sharpness";
}
//check brightness
if ((face.Quality.Brightness != null) && (face.Quality.Brightness > 0.61F) && (face.Quality.Brightness < 0.97F))
{
FacialAnalysisData += "\n✔ Picture is acceptable brightness";
}
else
{
FacialAnalysisData += "\n❌ Picture is acceptable brightness";
}
}
await DisplayAlert("Analysis Results", FacialAnalysisData, "OK");
}
catch (Exception exception)
{
Console.WriteLine(exception.Message);
}
}