News classification
Contact us
- Add: No. 9, North Fourth Ring Road, Haidian District, Beijing. It mainly includes face recognition, living detection, ID card recognition, bank card recognition, business card recognition, license plate recognition, OCR recognition, and intelligent recognition technology.
- Tel: 13146317170 廖经理
- Fax:
- Email: 398017534@qq.com
Target detection method based on AI depth learning
Target detection method based on AI depth learning
introduction
The task of target detection is to find out all the interesting objects in the image, and to affirm their position and size, which is one of the central problems in the category of machine vision. Because all kinds of objects have different appearance, shape, posture, interference with imaging, shade and occlusion, the purpose of detection is to be the most combat oriented in the category of machine vision. This paper will analyze the classic tasks of machine vision in Object Detection. If you have different ideas about the content of the text, send it to the SIGAI public address and send it to us for discussion.
What is the purpose test?
The task of target detection is to find out all the interesting objects in the image, and to affirm their position and size, which is one of the central problems in the category of machine vision. Because all kinds of objects have different appearance, shape, posture, interference with imaging, shade and occlusion, the purpose of detection is to be the most combat oriented in the category of machine vision.
There are four main tasks in computer vision about image recognition:
Classification -Classification: what is the "what?" The problem is to give a picture or a video to determine what kind of purpose it contains.
Locate -Location: handle "where?" The problem is to locate the location of this purpose.
Detect -Detection: "what is it?" Where is it? " The problem is locating the location of the purpose and knowing what the object is.
Partition -Segmentation: the partition (Instance-level) and the scene division (Scene-level), which are divided into instance, and deal with "which object or scene each pixel belongs to".
Aim to detect the central problem to be dealt with
In addition to image classification, the central problem of target detection is:
The 1. object may be present at any position of the image.
2. purposes have different sizes.
3. purposes may have a variety of shapes.
If a rectangle is used to define the destination, the rectangle has different aspect ratio. Because of the different aspect ratio, the classic sliding window and image scaling scheme is too costly to deal with the general purpose detection problem.
Application of target detection
Objective detection is applied in many categories. Among them are widely studied, such as face detection, pedestrian detection, vehicle detection and other important purposes. Face detection has been briefly introduced in an article on the history of face recognition algorithm in SIGAI. Later, we will write a summary of this issue.
Pedestrian detection
Pedestrian detection plays an important role in video surveillance, traffic statistics and autopilot.
Vehicle detection
Vehicle detection plays an important role in intelligent transportation, video surveillance and automatic driving. Car traffic statistics, automatic analysis of vehicle violations, such as automatic analysis, in automatic driving, the first to deal with the problem is where to affirm the road, there are cars, people or obstacles around.
The task of target detection is to find out all the interesting objects in the image, and to affirm their position and size, which is one of the central problems in the category of machine vision. Because all kinds of objects have different appearance, shape, posture, interference with imaging, shade and occlusion, the purpose of detection is to be the most combat oriented in the category of machine vision. This paper will analyze the classic tasks of machine vision in Object Detection. If you have different ideas about the content of the text, send it to the SIGAI public address and send it to us for discussion.
What is the purpose test?
The task of target detection is to find out all the interesting objects in the image, and to affirm their position and size, which is one of the central problems in the category of machine vision. Because all kinds of objects have different appearance, shape, posture, interference with imaging, shade and occlusion, the purpose of detection is to be the most combat oriented in the category of machine vision.
There are four main tasks in computer vision about image recognition:
Classification -Classification: what is the "what?" The problem is to give a picture or a video to determine what kind of purpose it contains.
Locate -Location: handle "where?" The problem is to locate the location of this purpose.
Detect -Detection: "what is it?" Where is it? " The problem is locating the location of the purpose and knowing what the object is.
Partition -Segmentation: the partition (Instance-level) and the scene division (Scene-level), which are divided into instance, and deal with "which object or scene each pixel belongs to".
Aim to detect the central problem to be dealt with
In addition to image classification, the central problem of target detection is:
The 1. object may be present at any position of the image.
2. purposes have different sizes.
3. purposes may have a variety of shapes.
If a rectangle is used to define the destination, the rectangle has different aspect ratio. Because of the different aspect ratio, the classic sliding window and image scaling scheme is too costly to deal with the general purpose detection problem.
Application of target detection
Objective detection is applied in many categories. Among them are widely studied, such as face detection, pedestrian detection, vehicle detection and other important purposes. Face detection has been briefly introduced in an article on the history of face recognition algorithm in SIGAI. Later, we will write a summary of this issue.
Pedestrian detection
Pedestrian detection plays an important role in video surveillance, traffic statistics and autopilot.
Vehicle detection
Vehicle detection plays an important role in intelligent transportation, video surveillance and automatic driving. Car traffic statistics, automatic analysis of vehicle violations, such as automatic analysis, in automatic driving, the first to deal with the problem is where to affirm the road, there are cars, people or obstacles around.