AWS SDK を使用して、リファレンスイメージと比較した Amazon Rekognition コレクション内の顔を検索する

次のコード例は、リファレンスイメージと比較した Amazon Rekognition コレクション内の顔を検索する方法を示しています。

詳細については、「顔を検索する (イメージ)」を参照してください。

.NET

AWS SDK for .NET

注記

他にもありますGitHub。用例一覧を検索し、AWS コード例リポジトリでの設定と実行の方法を確認してください。

    using System;
    using System.Threading.Tasks;
    using Amazon.Rekognition;
    using Amazon.Rekognition.Model;

    /// <summary>
    /// Uses the Amazon Rekognition Service to search for images matching those
    /// in a collection. The example was created using the AWS SDK for .NET
    /// version 3.7 and .NET Core 5.0.
    /// </summary>
    public class SearchFacesMatchingImage
    {
        public static async Task Main()
        {
            string collectionId = "MyCollection";
            string bucket = "bucket";
            string photo = "input.jpg";

            var rekognitionClient = new AmazonRekognitionClient();

            // Get an image object from S3 bucket.
            var image = new Image()
            {
                S3Object = new S3Object()
                {
                    Bucket = bucket,
                    Name = photo,
                },
            };

            var searchFacesByImageRequest = new SearchFacesByImageRequest()
            {
                CollectionId = collectionId,
                Image = image,
                FaceMatchThreshold = 70F,
                MaxFaces = 2,
            };

            SearchFacesByImageResponse searchFacesByImageResponse = await rekognitionClient.SearchFacesByImageAsync(searchFacesByImageRequest);

            Console.WriteLine("Faces matching largest face in image from " + photo);
            searchFacesByImageResponse.FaceMatches.ForEach(face =>
            {
                Console.WriteLine($"FaceId: {face.Face.FaceId}, Similarity: {face.Similarity}");
            });
        }
    }

• API の詳細については、AWS SDK for .NETAPI SearchFacesByImageリファレンスのを参照してください。

Java

SDK for Java 2.x

注記

他にもありますGitHub。用例一覧を検索し、AWS コード例リポジトリでの設定と実行の方法を確認してください。

    public static void searchFacebyId(RekognitionClient rekClient,String collectionId, String faceId) {

        try {
            SearchFacesRequest searchFacesRequest = SearchFacesRequest.builder()
                .collectionId(collectionId)
                .faceId(faceId)
                .faceMatchThreshold(70F)
                .maxFaces(2)
                .build();

            SearchFacesResponse imageResponse = rekClient.searchFaces(searchFacesRequest) ;
            System.out.println("Faces matching in the collection");
            List<FaceMatch> faceImageMatches = imageResponse.faceMatches();
            for (FaceMatch face: faceImageMatches) {
                System.out.println("The similarity level is  "+face.similarity());
                System.out.println();
            }

        } catch (RekognitionException e) {
            System.out.println(e.getMessage());
            System.exit(1);
        }
    }

• API の詳細については、AWS SDK for Java 2.xAPI SearchFacesByImageリファレンスのを参照してください。

Python

SDK for Python (Boto3)

注記

他にもありますGitHub。用例一覧を検索し、AWS コード例リポジトリでの設定と実行の方法を確認してください。

class RekognitionCollection:
    """
    Encapsulates an Amazon Rekognition collection. This class is a thin wrapper
    around parts of the Boto3 Amazon Rekognition API.
    """
    def __init__(self, collection, rekognition_client):
        """
        Initializes a collection object.

        :param collection: Collection data in the format returned by a call to
                           create_collection.
        :param rekognition_client: A Boto3 Rekognition client.
        """
        self.collection_id = collection['CollectionId']
        self.collection_arn, self.face_count, self.created = self._unpack_collection(
            collection)
        self.rekognition_client = rekognition_client

    @staticmethod
    def _unpack_collection(collection):
        """
        Unpacks optional parts of a collection that can be returned by
        describe_collection.

        :param collection: The collection data.
        :return: A tuple of the data in the collection.
        """
        return (
            collection.get('CollectionArn'),
            collection.get('FaceCount', 0),
            collection.get('CreationTimestamp'))

    def search_faces_by_image(self, image, threshold, max_faces):
        """
        Searches for faces in the collection that match the largest face in the
        reference image.

        :param image: The image that contains the reference face to search for.
        :param threshold: The match confidence must be greater than this value
                          for a face to be included in the results.
        :param max_faces: The maximum number of faces to return.
        :return: A tuple. The first element is the face found in the reference image.
                 The second element is the list of matching faces found in the
                 collection.
        """
        try:
            response = self.rekognition_client.search_faces_by_image(
                CollectionId=self.collection_id, Image=image.image,
                FaceMatchThreshold=threshold, MaxFaces=max_faces)
            image_face = RekognitionFace({
                'BoundingBox': response['SearchedFaceBoundingBox'],
                'Confidence': response['SearchedFaceConfidence']
            })
            collection_faces = [
                RekognitionFace(face['Face']) for face in response['FaceMatches']]
            logger.info("Found %s faces in the collection that match the largest "
                        "face in %s.", len(collection_faces), image.image_name)
        except ClientError:
            logger.exception(
                "Couldn't search for faces in %s that match %s.", self.collection_id,
                image.image_name)
            raise
        else:
            return image_face, collection_faces

• API の詳細については、「AWSSDK for Python (Boto3) API リファレンス」のを参照してくださいSearchFacesByImage。