口罩识别

1、定义：指定区域内人员佩戴口罩检测
2、使用场景：厨房、食堂、冷冻仓库、市场等各类食品生产场所
3、口罩识别类型：
1）支持：一次性医用口罩、红色口罩，粉色口罩，黑色口罩，白色口罩
2）不支持：人脸图案或者卡通图案口罩、防飞沫食堂口罩、和面部颜色相近的口罩
4、人脸姿态：
支持：人脸角度为正脸或小角度侧脸（左右角度小于60度）、轻微低头（俯仰角小于30度），人脸五官清晰可见
精度对应场景或者限制条件
）图片分辨率：1080p及以上。
2）设备/镜头：支持RGB可见光彩色图像；不支持红外图像。
3）成像条件：画面清晰，无明显拖尾、抖动、异常颗粒、偏色等模糊情况。画面曝光合适，无过曝引起的高光溢

1. Definition: Mask wearing detection within a specified area
2. Application Scenarios: Various food production sites including kitchens, canteens, cold storage, markets, and other similar places
3. Mask Recognition Types:
1) Supported types: Disposable medical masks, red masks, pink masks, black masks, and white masks
2) Unsupported types: Masks with facial or cartoon patterns, anti-spray canteen masks, and masks with colors similar to facial skin tone
4. Facial Posture Requirements:
Supported conditions: Frontal face or small-angle profile (left/right angle less than 60°), slight head lowering (pitch angle less than 30°), with clear facial features visible.
Precision-related scenarios and constraints:
1) Image Resolution: 1080p and above
2) Equipment/Lens: Supports RGB visible light color images; does not support infrared images
3) Imaging Conditions: The image must be clear, free of obvious trailing artifacts, jitter, abnormal grain, color cast and other blurring issues. The exposure should be appropriate, without highlight overflow caused by overexposure.

搜集汇总

背景与挑战

背景概述

该数据集是一个口罩识别算法模型，支持检测多种颜色的一次性医用口罩，适用于厨房、食堂等食品生产场所。识别要求包括人脸姿态为正脸或小角度侧脸，且需要1080p及以上分辨率的RGB可见光彩色图像。