作品說明

利用AI辨識系統與生物特徵辨識，提供各公司在疫情期間能有效率且無接觸的進行身分辨識的門禁系統。為了能讓公司員工能帶著口罩進行生物特徵辨識以利防疫，我們採用耳朵作為我們的辨識生物特徵。 由公司取得員工耳朵特徵圖片，輸入進以訓練好之模型，透過OpenCv cascade使門禁系統上的鏡頭自動擷取出人耳的特徵已用來辨識，最後再透過以訓練好的AI model來進行耳朵與資料庫中的員工耳朵照片進行比對，已確認是否是該公司的員工並顯示該員工的資料，再回傳結果給後方的閘門使其開門。當辨識到非資料庫中的人物時，會回傳訊息給管理者，並將閘門警報響起通知管理員查看。 透過鏡頭辨識取代證件辨識或磁卡辨識可以減少在疫情期間的接觸，降低疾病傳染風險，並提高時間效率。 以下是我們想要解決的問題: 1.減少接觸辨識方式的可能，避免病毒的傳播 2.精準的辨識，避免外人隨意進入公司 3.證件辨識所照成時間上的浪費 By utilizing the AI image recognition system, our project will be used to provide companies with an effective and contact-free access control system for identity identification during the epidemic. In order to allow our employees to wear masks for biometric identification, ears will be used for biometric recognition. Through AI image recognition, the lens on the access control system automatically extracts the features of human ears for identification. then, through the OpenCv cascade, the photos of the ears are compared with the photos of employees' ears in the database to confirm whether they are employees of the company. The information will therefore be sent back to the gate.When a person who is not in the database is identified, a message is sent back to the administrator and the alarm is sounded to inform the administrator. Image recognition instead of id identification or RFID identification can reduce contact during an epidemic, reduce the risk of disease transmission, and improve time efficiency. Here are the problems we want to solve: 1. Reduce the possibility of contact identification to avoid the spread of the virus 2. Accurate identification to prevent outsiders from entering the company at will 3. Reduce the time wasted by identification