基于卷積神經(jīng)網(wǎng)絡(luò)的包裝盒缺陷檢測(cè)理論分析研究

打開(kāi)文本圖片集

摘要：為了檢測(cè)出包裝盒表面缺陷部位，提高包裝盒表面缺陷檢測(cè)的準(zhǔn)確率，采用深度學(xué)習(xí)中的卷積神經(jīng)網(wǎng)絡(luò)對(duì)包裝盒表面缺陷進(jìn)行檢測(cè)。本文介紹卷積神經(jīng)網(wǎng)絡(luò)的基本理論，以及卷積神經(jīng)網(wǎng)絡(luò)常見(jiàn)的網(wǎng)絡(luò)結(jié)構(gòu)，并且對(duì)常見(jiàn)的神經(jīng)網(wǎng)絡(luò)進(jìn)行歸納總結(jié)。

關(guān)鍵詞：缺陷檢測(cè)；深度學(xué)習(xí)；卷積神經(jīng)網(wǎng)絡(luò)；包裝盒

中圖分類號(hào)：TB48 文獻(xiàn)標(biāo)識(shí)碼：A 文章編號(hào)：1400 （2022） 09-0032-04

Theoretical Analysis and Research on Surface Defect Detection of Packaging Box Based on Convolution Neural Network

WANG Fu-hao， CAI Ji-fei， SHI Mo-yan， YIN Tong， LUO Jian-qing（Beijing Institute of Graphic Communication， Beijing 102600， China）

Abstract： In order to detect the surface defects of the packaging box and improve the accuracy of the surface defect detection of the packaging box， the convolution neural network in deep learning is used to detect the surface defects of the packaging box. This paper introduces the basic theory of convolutional neural network and the common network structure of convolutional neural network， and summarizes the common neural networks.

Key words： defect detection； deep learning； convolutional neural network； packing box

缺陷檢測(cè)是非常重要的環(huán)節(jié)，尤其在印刷、包裝、紡織等領(lǐng)域有著非常廣泛的應(yīng)用。（剩余3188字）