org.springframework.cloud.fn.object.detection

Class ObjectDetectionService

java.lang.Object

org.springframework.cloud.fn.object.detection.ObjectDetectionService

public class ObjectDetectionService
extends Object

Convenience class that leverages the the ObjectDetectionInputConverter, ObjectDetectionOutputConverter and TensorFlowService in combination fromMemory the Tensorflow Object Detection API (https://github.com/tensorflow/models/tree/master/research/object_detection) models for detection objects in input images. All pre-trained models (https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md) and labels are supported. You can download pre-trained models directly from the zoo: https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md Just use the URI notation: (zoo model tar.gz url)#(name of the frozen model file name). To speedup the bootstrap performance you should consider downloading the models locally and use the file:/"path to my model" URI instead! The object category labels for the pre-trained models are available at: https://github.com/tensorflow/models/tree/master/research/object_detection/data Use the labels applicable for the model. Also, for performance reasons you may consider to download the labels and load them from file: instead.

Author:

Christian Tzolov

Field Summary

Fields

Modifier and Type	Field and Description
static List<String>	FETCH_NAMES Default list of fetch names for Box models.
static List<String>	FETCH_NAMES_WITH_MASKS Default list of fetch names for mask supporting models.

Constructor Summary

Constructors

Constructor and Description
ObjectDetectionService()
ObjectDetectionService(ObjectDetectionInputConverter inputConverter, ObjectDetectionOutputConverter outputConverter, org.springframework.cloud.fn.common.tensorflow.deprecated.TensorFlowService tensorFlowService) Generic constructor thea allow the converter to be pre-configured before passed to the service.
ObjectDetectionService(String modelUri, String labelsUri, float confidence, boolean withMasks, boolean cacheModel) Convenience constructor that would initialize all necessary internal components.

Method Summary

Modifier and Type	Method and Description
List<ObjectDetection>	detect(BufferedImage image, String format) Detects objects in a single BufferedImage.
List<ObjectDetection>	detect(byte[] image) Detects objects from a single input image encoded as byte array.
List<List<ObjectDetection>>	detect(byte[][] batchedImages) Uses detects objects from a batch of input images encoded as byte array.
List<ObjectDetection>	detect(String imageUri) Detects objects in a single input image identified by its URI.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

FETCH_NAMES

public static List<String> FETCH_NAMES

Default list of fetch names for Box models.

FETCH_NAMES_WITH_MASKS

public static List<String> FETCH_NAMES_WITH_MASKS

Default list of fetch names for mask supporting models.

Constructor Detail

ObjectDetectionService

public ObjectDetectionService()

ObjectDetectionService

public ObjectDetectionService(String modelUri,
                              String labelsUri,
                              float confidence,
                              boolean withMasks,
                              boolean cacheModel)

Convenience constructor that would initialize all necessary internal components.

Parameters:

modelUri - URI of the pre-trained, frozen Tensorflow model.

labelsUri - URI of the pre-trained category labels.

confidence - Confidence threshold. Only objects detected wth confidence above this threshold will be returned.

withMasks - If a Mask model is selected then you can use this flag to extract the instance segmentation masks as well.

ObjectDetectionService

public ObjectDetectionService(ObjectDetectionInputConverter inputConverter,
                              ObjectDetectionOutputConverter outputConverter,
                              org.springframework.cloud.fn.common.tensorflow.deprecated.TensorFlowService tensorFlowService)

Generic constructor thea allow the converter to be pre-configured before passed to the service.

Parameters:

inputConverter - Converter from byte array to object detection input image tensor

outputConverter - Covets the object detection output tensors into ObjectDetection list

tensorFlowService - Java tensorflow runner instance

Method Detail

detect

public List<ObjectDetection> detect(String imageUri)

Detects objects in a single input image identified by its URI.

Parameters:

imageUri - input image's URI

Returns:

Returns a list of ObjectDetection domain objects representing detected objects

detect

public List<ObjectDetection> detect(BufferedImage image,
                                    String format)

Detects objects in a single BufferedImage.

Parameters:

image - Input image to detect objects from.

format - Image format (e.g. jpg, png ...) to use when converting the buffer into byte array.

Returns:

Returns a list of ObjectDetection domain objects representing detected objects in the input image

detect

public List<ObjectDetection> detect(byte[] image)

Detects objects from a single input image encoded as byte array.

Parameters:

image - Input image encoded as byte array

Returns:

Returns a list of ObjectDetection domain objects representing detected objects in the input image

detect

public List<List<ObjectDetection>> detect(byte[][] batchedImages)

Uses detects objects from a batch of input images encoded as byte array.

Parameters:

batchedImages - Batch of input images encoded as byte arrays. First dimension is the batch size and second the image bytes.

Returns:

Returns list of lists. For every input image in the batch a list of ObjectDetection domain objects representing detected objects in the input image.