Introduction to OpenCV 3.x (with Java)

Computer Vision
OpenCV
Luigi De Russis
Politecnico di Torino
Dipartimento di Automatica e Informatica (DAUIN)
Torino - Italy
luigi.derussis@polito.it
This work is licensed under the Creative Commons (CC BY-NC-SA)
License. To view a copy of this license, visit
http://creativecommons.org/licenses/by-nc-sa/4.0/

What is OpenCV?
 Computer Vision library
 Open Source (BSD License)
 http://opencv.org
 Originally developed by Intel
 With more than 2500 optimized algorithms
 Supports a lot of different languages
 C, C++, Python, and Java
 but is written natively in C++
 Cross platform
 also available for Android and iOS
10/18/2016 Introduction to OpenCV 2

What is OpenCV?
SSE, NEON, IPP, OpenCL, CUDA, etc.
core, imgproc, objdetect, etc.
face, text, etc.
Bindings: Python,
Java, …
Sample, tutorials
OpenCV
OpenCV
contrib
OpenCV
HAL

What is used for?
 Human-Computer Interaction (HCI)
 Object Identification
 Object Recognition
 Face Recognition
 Gesture Recognition
 Motion Tracking
 Image Processing
 Mobile Robotics
 … and so on

Why OpenCV (for this course)?
 We looked for a software/library:
 that covers most of Computer Vision techniques
 widely used or de-facto standard
 free/open, possibly
 in a language known by most students
 in which a GUI can be created
 able to handle real-time image processing

Alternatives?
vs.

OpenCV: pros
 Specificity
 OpenCV was made for computer vision/image
processing
 Matlab is quite generic
 Speed
 30+ frames processed per seconds in real time image
processing with OpenCV
 around 4-5 frames processed per seconds in real time
image processing with Matlab
 Efficiency
 Matlab needs more system resources than OpenCV

OpenCV: cons
 Easy of use
 Matlab won hands down!
 Integrated Development Environment (IDE)
 you can use Eclipse, Netbeans, Visual Studio, Qt,
XCode, … even a simple text editor for OpenCV
 Matlab has is own IDE
 Memory management

OpenCV: pros (final)
 Price (!)
 OpenCV Wrappers
 Emgu CV, ruby-opencv…
 More similar to industry-level frameworks

OpenCV in C++: why not?
 The OpenCV APIs have few capabilities for
user interfaces
 i.e., you can open a window with an image inside.
Do you want a button? You cannot have it!
 to have a “decent” user interface you need to
install “something else” (e.g., Qt for C++)…
 … and then compile OpenCV with Qt support
 C++ learning curve is quite… steep
 source: experiences from teaching this course in
the first years

 OpenCV C++ APIs exist since OpenCV 1.0
 they dramatically changed starting from
OpenCV 2.0
 it is possible, but not recommended, to “mix”
different OpenCV version
 typically, it is a good way to get into trouble!

 It is possible to use OpenCV C APIs together
with C++ APIs
 typically it is a good way to get into trouble!
 OpenCV 3.x is mostly compatible with
OpenCV 2.x
 but OpenCV 1.x API is deprecated and partially
removed

OpenCV in Java: why?
 OpenCV provides also Java APIs
 nothing to compile (on Windows)
 "easy" installation on macOS thanks to
Homebrew
 they works on Android, too
 Java 7/8 has a first-grade user interface
 not Swing, but JavaFX
 https://docs.oracle.com/javase/8/javase-
clienttechnologies.htm

 Almost everybody here should have some
(basic) experience with Java
 Performances between the C++ APIs and the
Java APIs are comparable

 The OpenCV Java APIs are really similar to the
C++ version
 knowledge can be transferred!
 examples:
 to store an image in memory both use the Mat object
 to write an image on disk both use the imwrite() method
 Something changes a bit
 examples:
 CV_RGB2GRAY (C++) becomes COLOR_RGB2GRAY (Java)
 Point, Point2d, Point2f (C++) becomes 3 overloaded
constructor of Point (Java)

OPENCV IN JAVA
Getting started…

OpenCV 3.x
 It is modular and extendible
 It includes mature and clean algorithms that
are fully supported
 A separate contribution repository is for new
algorithms that people want to share
OpenCV 3.x
contributions
OpenCV
3.x
OpenCV 2.x

Main Modules
 core
 basic structures and algorithms
 imgproc
 image processing algorithms (such as image filtering,
geometrical image transformations, histograms, etc.)
 videoio
 media I/O
OpenCV has a modular structure:
each package includes several shared or static libraries
New

Main Modules
 video
 video analysis (such as motion estimation and object
tracking)
 imgcodecs
 image file reading and writing
 calib3d
 camera calibration and 3D reconstruction
 features2d
 2D features framework (feature detectors, descriptors, and
descriptor matchers)
 objdetect
 detection of objects and other items (e.g., faces, eyes,
mugs, people, …)
New

Main Modules
 highgui
 provides basic UI capabilities (in C, C++, and Python)
 ml
 machine learning classes used for statistical
classification, regression and clustering of data
 photo
 computational photography
 cuda*
 various modules with CUDA-optimized algorithms
New
Not available in Java

Data Structures
 We speak about Java APIs
 All the OpenCV classes and methods are placed
into the org.opencv.* packages
 Mat
 the primary image structure in OpenCV 2.x/3.x
 overcomes the “old” IplImage/CvMat problems
(OpenCV 1.x/C API)
 automatic memory management (more or less in C++)
 two data parts:
 matrix header (contains information about the matrix)
 a pointer to the matrix containing the pixel values

Duplicate Mat objects…
Mat first = Mat.eye(3, 3, CvType.CV_8U1C);
1 0 0
0 1 0
0 0 1
Header
first

Mat second = first;
1 0 0
0 1 0
0 0 1
Header
first
Header
second

Mat second = first;
first.put(0, 0, 2, 2, 2);
Header
first
Header
second
2 2 2
0 1 0
0 0 1

Mat second = new Mat();
second = first.clone(); // first.copyTo(second);
first.put(0, 0, 2, 2, 2);
2 2 2
0 1 0
0 0 1
Header
first
Header
second
1 0 0
0 1 0
0 0 1

Data Structures
 Point
 2D point
 defined by x, y coordinates
 Point first = new Point(2, 3);
 Size
 2D size structure
 specify the size (width and height) of an image
or rectangle
 Rect
 2D rectangle object

Basic Drawing Operations
 Imgproc.circle()
 draws a simple or filled circle with a given center and radius
on a given image
 Imgproc.line()
 draws a line between two point in the given image
 Imgproc.ellipse()
 draws an ellipse outline, a filled ellipse, an elliptic arc, a filled
ellipse sector, …
 Imgproc.rectangle()
 draws a rectangle outline or a filled rectangle
 note that negative thickness will fill the rectangle

Basic Image I/O
 Imgcodecs.imread()
 loads an image from file and return the corresponding
Mat object
Mat Imgcodecs.imread(String filename,
int flags)
 Imgcodecs.imwrite()
 save an image on disk
bool Imgcodecs.imwrite(String filename,
Mat img, MatOfInt params)

Color Spaces
 Imgproc.cvtColor
 converts an input image from one color space to another
 examples:
 cvtColor(src, dest, Imgproc.COLOR_RGB2GRAY);
 cvtColor(src, dest, Imgproc.COLOR_HSV2BGR);
 cvtColor(src, dest, Imgproc.COLOR_RGB2BGR);
 Important! Images in OpenCV uses BGR instead of
RGB

Getting started (without going crazy)
LET’S GET PRACTICAL!
10/18/2016 30 Introduction to OpenCV

How can we use OpenCV?
 LABINF:
 already installed under Windows
 Java 8 and OpenCV version 3.1
 Eclipse (Neon) is the IDE
 At home:
 we recommend Eclipse Neon, Java 8 and OpenCV
3.1
 feel free to use any Java IDE you like
 Installation:
 http://opencv-java-
tutorials.readthedocs.io/en/latest/01-installing-opencv-
for-java.html

What if I got problems?
 Small problems
 drop me a line
 luigi.derussis@polito.it
 Normal problems
 come to office hours
 please send an e-mail beforehand
 Enormous problems
 pray?
 no, seriously, we can schedule an extra “lesson”
Awesome student to me
Hi,
[…] I’m using “cvtColor(image, gray, COLOR_BGR2GRAY);” but it give this exception: […]
Can you help me?
Regards,
…
Problems with JavaFX and a gray scale image

 Small problems
 drop me a line
 Normal problems
 pray?
Not-So-Awesome student to me
Hi,
[…] I followed the guide for installing OpenCV on my Mac but I have an error after step 3. Can
we meet on next Wednesday to solve the problem?
Thanks!
Regards,
…
OpenCV installation

 Small problems
 drop me a line
 Normal problems
 pray?
Good student to me
Hi,
[…] I see the solution of Exercise 2.1 but I don’t understand the following expressions:
- main();
- System.out.println();
- @Override.
Can you explain to me what they are?
Regards,
…
Help with OpenCV

An e-mail not to be sent!

Useful Resources…
 OpenCV Wiki
 https://github.com/Itseez/opencv/wiki
 OpenCV Official Documentation
 http://docs.opencv.org/3.1.0/
 User Q&A forum
 http://answers.opencv.org/questions/
 OpenCV Javadocs
 http://docs.opencv.org/java/3.1.0/
 JavaFX Documentation
 https://docs.oracle.com/javase/8/javase-
clienttechnologies.htm

Useful Resources…
 Daniel Lélis Baggio, OpenCV 3.0 Computer
Vision with Java, PACKT Publishing, 2015,
ISBN 978-1-78328-397-2
 Gary Bradski, Adrian Kaehler, Learning
OpenCV 3 - Computer Vision in C++ with the
OpenCV Library, O'Reilly Media, TBP, ISBN
978-1-4919-3799-0
 Other books:
 http://opencv.org/books.html

Put everything together
DEMO HOUR
10/18/2016 38 Introduction to OpenCV

GUI? No, thanks
 Write a simple OpenCV program to:
1. create and print a 3x3 identity matrix
2. read an image from disk
3. convert the image in grayscale
4. save the grayscale image back on disk

OpenCV meets JavaFX (I)
 Write a graphical application with OpenCV to:
1. read an image from disk
2. convert it in grayscale
3. show the result on screen

OpenCV meets JavaFX (II)
 Write a graphical application with OpenCV to:
1. acquire the video stream from the webcam
2. convert it in grayscale
3. show the result of the acquisition on screen

License
18/10/2016 Computer Vision - Introduction 1 42
This work is licensed under the Creative Commons “Attribution-
NonCommercial-ShareAlike International (CC BY-NC-SA 4.0)”
License.
You are free to:
 Share - copy and redistribute the material in any medium or format
 Adapt - remix, transform, and build upon the material
for any purpose, even commercially.
Under the following terms:
 Attribution - You must give appropriate credit, provide a link to the
license, and indicate if changes were made. You may do so in any
reasonable manner, but not in any way that suggests the licensor
endorses you or your use.
 Noncommercial - You may not use the material for commercial
purposes.
 Share Alike - If you remix, transform, or build upon the material, you
must distribute your contributions under the same license as the
original.

Introduction to OpenCV 3.x (with Java)

Recomendados

Recomendados

Mais conteúdo relacionado

Mais procurados

Mais procurados (20)

Destaque

Destaque (6)

Semelhante a Introduction to OpenCV 3.x (with Java)

Semelhante a Introduction to OpenCV 3.x (with Java) (20)

Mais de Luigi De Russis

Mais de Luigi De Russis (20)

Último

Último (20)

Introduction to OpenCV 3.x (with Java)