Computer vision is a simulation of biological vision using computers and related equipment. Its main task is to obtain 3D information of the corresponding scene by processing the collected pictures or videos, just like humans and many other creatures do every day.
Computer vision is a challenging and important area of research in both engineering and science. Computer vision is a comprehensive subject, which has attracted researchers from various disciplines to participate in its research. These include computer science and engineering, signal processing, physics, applied mathematics and statistics, neurophysiology and cognitive science, among others.
So first of all - why learn computer vision?
From his application, you can clearly see his importance!
Examples of computer vision applications include systems for:
(1) Controlling a process, for example, an industrial robot;
(2) Navigation, for example, by autonomous vehicles or mobile robots;
(3) Detected events, such as video surveillance and people counting;
(4) organizing information, for example, indexed databases for images and image sequences;
(5) Modeling objects or environments, such as medical image analysis systems or terrain models;
(6) Interaction, for example, when input to a device, for computer-human interaction;
(7) Automatic detection, for example, in manufacturing applications.
(8) Autonomous car driving
(9) Facial recognition
1. Image classification
definition
Image classification is an image processing method that distinguishes different types of objects according to the different characteristics reflected in the image information. It uses computer to carry out quantitative analysis on images, and classifies each pixel or area in the image or image into one of several categories to replace human visual interpretation.
Classification
1. Index technology based on color features
2. Texture-based image classification technology
3. Shape-based image classification technology
4. Image classification technology based on spatial relationship
2. Target detection
Object detection, also called object extraction, is an image segmentation based on the geometric and statistical characteristics of the object. It combines target segmentation and recognition into one, and its accuracy and real-time performance are an important capability of the whole system.
As an important branch of computer vision, the task of target detection is to find the target category and target location in an image or video. Different from image classification, object detection focuses on object search, and the detected object must have a fixed shape and outline; while image classification can be any object including objects, attributes, and scenes. Target detection has achieved very remarkable results in the fields of face recognition and automatic driving. The classic detection models include YOLOV3, SSD and Faster RCNN.
It combines target segmentation and recognition into one, and its accuracy and real-time performance are an important capability of the whole system. Especially in complex scenes, when multiple targets need to be processed in real time, automatic target extraction and recognition is particularly important.
With the development of computer technology and the wide application of computer vision principles, the use of computer image processing technology to track targets in real time is becoming more and more popular. Dynamic real-time tracking and positioning of targets is important in intelligent transportation systems, intelligent monitoring systems, and military target detection. It has a wide range of application value in the positioning of surgical instruments in medical navigation surgery.
3. Target tracking
Object tracking aims to track the motion of objects in a video.
Usually, the position of the object in the first frame of the video is given in the form of a bounding box, and we need to predict the bounding box of the object in other frames. Target tracking is similar to target detection, but the difficulty of target tracking is that you don't know in advance what the target to track is, so you can't collect enough training data in advance to train a special detector.
The discriminative method is more powerful and accurate. It can be used to distinguish the difference between the subject and the background and has become the preferred tracking method. It is also known as Tracking-by-Detection and falls into the same category as deep learning.
4. Semantic Segmentation
Semantic segmentation is to classify each pixel in an image, and is currently widely used in medical images and driverless vehicles . Judging from the papers in recent years, this field is mainly divided into supervised semantic segmentation , unsupervised semantic segmentation , and video semantic segmentation .
Segmentation is an important part of computer vision, which divides the entire image into groups of pixels that can be labeled and classified. More specifically, semantic segmentation attempts to understand the role of each pixel in a given image. For example, just detecting a person or a car is not enough. You must also be able to tell where all boundaries are. To do such a depiction, we need dense pixel-wise predictions from the model.
5. Instance Segmentation
Instance segmentation is a higher level task combining object detection and semantic segmentation
Instance segmentation classifies all the different instance classes, e.g. label ten cars with ten different colors. In terms of classification, there is usually a main image and the goal is to determine what the image really is. However, to segment all instances, a more complex process is required. If we have a complex scene with many overlapping objects and various backgrounds, we have to classify all objects and determine their differences, boundaries, and relationships between them.
6. Image reconstruction
In layman's terms, you have an old photo and want to reconstruct the damaged picture in it, which is image reconstruction. In 2013, Jay Chou even spent 100 million Taiwan dollars, using the latest high-tech "virtual image" from the Hollywood special effects team. "Reconstruction Technology" invited Teresa Teng to sing with him, fulfilling his wish to sing with the late "Queen of Songs" Teresa Teng.
Computer vision is one of the most popular researches at present. It is a multidisciplinary research covering computer science (graphics, algorithms, theoretical research, etc.), mathematics (information retrieval, machine learning), engineering (robots, NLP, etc.) , biology (neuroscience) and psychology (cognitive science). Because computer vision represents a relative understanding of the visual environment and context, many scientists believe that research in this field will lay the foundation for the development of the artificial intelligence industry.
Organized a deep learning neural network + CV computer vision learning (two major frameworks pytorch/tensorflow + source code courseware notes) + NLP and other materials, with python foundation, image processing opencv\natural language processing, machine learning, mathematics foundation and other resource libraries, If you want to learn artificial intelligence or switch to a high-paying industry, college students are very practical. There is no routine to provide, +vx internal resources for free, artificial intelligence question bank, Dachang interview question study outline self-study course outline and 200GAI data large The gift package is free~
If you have any questions, you can ask me
Welcome to scan the code to get AI free video data
❶ Artificial intelligence courses and projects (including courseware source code) can be written into the resume of enterprise-level project practice
❷ High-quality must-read books on artificial intelligence ("Bible" flower books, etc.) + collection of artificial intelligence papers
❸ Tutorials and supporting zi materials of well-known domestic and foreign masters (Goddess Li Feifei, Wu Enda, Li Mu)
❹ Super detailed explanation of artificial intelligence learning path + system learning zi materials
❺ High-quality artificial intelligence resource website arrangement, artificial intelligence industry report