ỨNG DỤNG CÁC KỸ THUẬT THỊ GIÁC MÁY TÍNH TRONG XÁC ĐỊNH MỆNH GIÁ TIỀN VÀ PHÁT HIỆN TIỀN GIẢ

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Van Dung Hoang, Hoang Thanh Võ

Ngôn ngữ: vie

Ký hiệu phân loại:

Thông tin xuất bản: Tạp chí Khoa học và Công nghệ - Trường Đại học Quảng Bình, 2020

Mô tả vật lý: tr.18-26

Bộ sưu tập: Metadata

ID: 321098

Thêm vào giỏ Liên kết toàn văn

Advanced technology has played an important task in circulation of anti-counterfeit notes economy. It is essential that requires an efficient solution to detect fake banknotes. This paper proposes an approach for recognition of paper currency based fundamental image processing using deep learning for feature extraction and recognition. Deep neural network techniques have dramatically become the state of the art in image processing. The high capacity of traditional techniques on currency image dataset has been impeded because of varieties of the appearance of the banknotes. This paper focuses recognition face value and anti-counterfeit based on banknote appearance. The proposed method can be applied to recognize many kinds of the denomination or face values as well as the national currencies. The contribution studies a new approach based on sequential deep neural network and data augmentation for improving accuracy. The deep neural network is constructed using several inceptions with different parallel convolutional operations which support reducing consuming time. Image augmentation of training dataset generates a larger data enough for deep neural network learning. This proposed task is aimed to address the small data problem. It is utilized for enhancing the capabilities of deep learning. Experimental results illustrate that the proposed method is applicable to the real application with enhances performance to 97.85% accuracy rate.Công nghệ tiên tiến đã đóng một nhiệm vụ quan trọng việc nhận dạng tiền thật trong các nền kinh tế. Bài báo này đề xuất một cách tiếp cận để nhận dạng tiền giấy dựa trên kỹ thuật xử lý ảnh và trí tuệ nhân tạo. Kỹ thuật học sâu trong trí tuệ nhân tạo dùng để trích xuất đặc trưng và nhận dạng loại tiền. Các kỹ thuật truyền thống thường có độ chính xác không cao trong phân tích xử lý số liệu hình ảnh trong nhận dạng mẫu. Bài báo này tập trung đề xuất kiến trúc học sâu để nhận dạng tiền thật và xác định mệnh giá tiền, phát hiện tiền giả. Giải pháp đề xuất gồm các nội dung như sau: Thứ nhất, mạng neural sâu được xây dựng gồm lớp ảnh đầu vào, một số lớp ẩn tích chập kết nối tuần tự, tiếp theo là các inceptions kết nối kiểu đồ thị có hướng và sau cùng là một số lớp kết nối đầy đủ và lớp đầu ra là số mệnh giá tiền và tiền giả. Thứ hai, kỹ thuật tăng cường hình ảnh dựa vào các phương pháp biến đổi hình học, màu sắc nhằm tăng cương tập dữ liệu đảm bảo tính bao quát cho việc huấn luyện mạng neural sâu có độ chính xác cao. Chúng tôi cũng tiến hành tạo lập bộ dữ liệu thực nghiệm với 06 mệnh giá tiền với 9736 mẫu, tiền giả với 1083 mẫu và hình ảnh không phải tiền với 1548 mẫu. Giải pháp đề xuất được thực nghiệm, đánh giá trên tập dữ liệu mẫu đạt độ chính xác 97,85%.

1. Đọc trực tuyến

1. Đọc trực tuyến

Tạo bộ sưu tập với mã QR