Preface
It is false to say that there is no anxiety, because no matter it is now or in recent years, some people have long noticed that the era of barbaric growth in Android development has passed. In the past, the advantage was market demand. Few people had this technology, so when they seized the market, it was enough to basically meet the demand. But now, all kinds of APP are emerging one after another, and the quality of APP has become a new competitive standard. This also means that whether it is system design or performance optimization, the market has put forward higher requirements for Android developers. So, is Android really unpopular? In fact, it is not. There is still a big demand gap for senior level and above, and talents are very scarce. But junior developers are really not popular anymore. To put it frustratingly, the competitive advantage of junior scholars in the market is actually very small.
We can also understand the current market situation from the demand position of Android development.
Knowledge points in this article
The overall structure of ID card recognition
- Effect demonstration
- Application scenarios of ID card recognition
- Project core structure
Key technology of ID card recognition
- Principles of NDK development
- What is JNI
- OpenCV architecture system
- OCR technology architecture
ID card recognition project combat
- Key architecture
- Core steps and handwritten implementation
- IOS windows running synchronously
- Android architecture
1. What is ID card recognition
Key problem analysis: find the area where the number is located → get the number picture → identify the ID number
2. Core
OpenCV:
The full name of OpenCV is Open Source Computer Vision Library, which is a cross-platform computer vision library. IBM began research and development in 1999, using C++ to provide interface Mat data structure
Application areas:
Image processing. Face recognition. Gesture recognition. Human-computer interaction action recognition. Motion tracking. Object recognition image segmentation
Tess-two:
The full name of OCR is Optical Character Recognition is a process of analyzing and recognizing image files of text data to obtain text and layout information.
tess-two is a git branch of TesseraToolsForAndroid
Use characteristics:
1: Simple and easy to use
2: Open source and support offline use
3: Java api customized for the Android platform
NDK development::
3. Detailed architecture
problem:
1: What is a grayscale image
2: What is binarization and why?
3: What is contour detection
4: What is image expansion
5: Will image segmentation cause OOM?
6: How to perform OCR text training
7: What is the API for graphic recognition?
3.1 The core steps of image recognition-grayscale
3.2 The core steps of image recognition-binarization
3.3 The core steps of image recognition-contour detection
All continuous closures are framed by rectangles
3.4 The core steps of image recognition-expansion
3.5 The core steps of image recognition-contour screening
3.6 text recognition
OCR text sample → memorized text → straight text recognition
to sum up
How to upload pictures?
It can’t be said that I will call how the interface is implemented, although I also do it
To understand:
1: Understand what Http is, and know how http uploads pictures.
2: Using okhttp, how does retrofit realize this process?
Learning route + knowledge combing
It took a long time to organize this detailed brain map of the knowledge route. Of course, due to limited time and abilities, after all, the entire embedded system is too large. Including my embedded classmates, it is impossible to understand everything. Some things may not be covered, and the shortcomings are still small. Partners communicate and supplement together, improve and progress together.
This article is in an open source project: What about Android development? How to get a high salary in an interview has been included, including self-learning programming routes in different directions, interview question collection/face-to-face, and a series of technical articles, etc. The resources are continuously being updated...
%A6%82%E4%BD%95%E9%9D%A2%E8%AF%95%E6%8B%BF%E9%AB%98%E8%96%AA%EF%BC%81.md) Medium It has been included, including self-learning programming routes in different directions, interview question collection/face sutras, and a series of technical articles, etc. The resources are continuously being updated...
Share it here this time, see you in the next article .