Variants refer to homographs, code words, and special symbols that cybercriminals use to output non-compliant content in an attempt to deceive smart recognition engines.
Optical character recognition (OCR) is a technology that converts the text in various documents, newspapers, books, manuscripts, and other printed materials into image information through optical input methods such as scanning and then converts such information into computer-readable input through text recognition technology.
Deep learning is a form of machine learning that combines low-level characteristics to form more abstract high-level representation attribute categories or characteristics in order to discover the distributed characteristic representation of data. The motivation for studying deep learning is to establish a neural network that simulates the analysis and learning processes of human brain. It mimics how human brain interprets data such as images, sounds, and text.
Image Moderation System
Image Moderation System (IMS) can accurately recognize content in images that may be offensive, unsafe, or inappropriate. It allows you to configure an image blocklist to recognize custom types of content.
Application Programming Interface
Application programming interface (API) refers to the code that a software system provides for an application to call.
Developers can call a set of functional APIs without having to consider the underlying source code or understand the internal working mechanisms.
API includes Windows API and Linux API. The APIs provided by Tencent Cloud IMS generally refer to Linux API.
Accuracy is the proportion of accurately recognized positive samples in all recognized samples. For example, if the image analysis API recognizes 100 images, and the tags in 95 of them are recognized correctly, then the accuracy is 95%.