/computer-vision

Collection of Spring Cloud Stream App Starters for Computer Vision (CV) and Image Processing

Primary LanguageJava

Spring Cloud Stream ComputerVision Applications

Collection of Spring Cloud Stream App Starters for Computer Vision (CV) and Image Processing.

Different 3rd party libraries are leveraged, including OpenIMAJ, javaCV, Sarxos, Zxing and more

Pre-build (snapshot) versions of the apps are available through the test-libs-snapshot-local maven repository. First register the snapshot maven repository to your SCDF server (using maven.remote-repositories.repo3.url property):

 --maven.remote-repositories.repo3.url=http://repo.spring.io/test-libs-snapshot-local

Then use CV-SNAPSHOT-stream-applications-rabbit-maven to bulk import the listed app starters:

 dataflow:>app import --uri https://raw.githubusercontent.com/tzolov/computer-vision/master/CV-SNAPSHOT-stream-applications-rabbit-maven

App Starters List


webcam source

Webcam Source that allows reading video streams from integrated or USB-connected webcams. Reads the video stream from a webcam, extracts single video frames as images and feeds them to the outbound binder.


frame-grabber source

javaCV (OpenCV java wrapper) powered processor Reads a file or camera video stream, extracts single video frames as images and feeds them to the outbound binder. You can configure the source location size of the output images and the frame grabbing interval.


frame-recorder sink

Description: TODO


ocr-reader processor

Description: TODO


qr-reader processor

Multi-format 1D/2D barcode image reader processor. Based on ZXing ("Zebra Crossing") barcode scanning library for Java.


image-viewer sink

Java Swing/JFrame based, image and video viewer. Plots the inbound message’s image payload to a JFrame panel.


face-detection-mtcnn processor

Description: WIP


semantic-segmentation processor

Description: Real-time Semantic Image Segmentation processor with Tensorflow, Java and Spring Cloud Stream. Semantic Segmentation is the process of associating each pixel of an image with a class label, (such as flower, person, road, sky, ocean, or car).