Variety identification model for maize seeds using hyperspectral pixel-level information combined with convolutional neural network

نویسندگان

چکیده

çŽ‰ç±³ä½œä¸ºä¸­å›½é‡è¦ç²®é£Ÿä½œç‰©ï¼Œå“ç§ä¼—å¤šï¼Œæ˜“å‡ºçŽ°é”™åˆ†çŽ°è±¡ï¼Œå½±å“å†œä¸šå®‰å ¨å’Œç²®é£Ÿç”Ÿäº§ã€‚é’ˆå¯¹ä¼ ç»ŸåŸºäºŽå·ç§¯ç¥žç»ç½‘ç»œCNN(Convolutional Neural Networkï¼‰çš„é«˜å ‰è°±å›¾åƒä½œç‰©å“ç§è¯†åˆ«æ¨¡åž‹æ‰€éœ€å»ºæ¨¡æ ·æœ¬æ•°é‡å·¨å¤§çš„é—®é¢˜ï¼Œæå‡ºåŸºäºŽé«˜å ‰è°±åƒç´ çº§ä¿¡æ¯å’ŒCNNçš„çŽ‰ç±³ç§å­å“ç§è¯†åˆ«æ¨¡åž‹ã€‚é¦–å ˆï¼ŒèŽ·å–ä¸åŒå“ç§çŽ‰ç±³ç§å­åœ¨400—1000 nmèŒƒå›´å† çš„é«˜å ‰è°±å›¾åƒï¼Œæå–æ ·æœ¬å ¨éƒ¨åƒç´ çš„203ç»´å ‰è°±ä¿¡æ¯ï¼Œåˆ©ç”¨ä¸»æˆåˆ†åˆ†æžPCA(Principal Component Analysisï¼‰ç®—æ³•å°†å ‰è°±ç»´åº¦é™è‡³8ç»´ã€‚åœ¨å®žéªŒä¸­ï¼Œæ ·æœ¬çš„åƒç´ çº§å ‰è°±ä¿¡æ¯ï¼ˆå³ï¼šæ ·æœ¬çš„å çš„å ‰è°±ä¿¡æ¯ï¼‰é™¤åº”ç”¨äºŽCNN模型外,也应用于支持向量机(SVM)和K近邻分类(KNNï¼‰æ¨¡åž‹ä¸­ï¼Œç»“æžœè¡¨æ˜Žï¼šåœ¨ç›¸åŒæ¨¡åž‹ä¸­ï¼ŒåŸºäºŽåƒç´ ‰è°±ä¿¡æ¯æ¯”åŸºäºŽç±³ç²’çº§å ‰è°±ä¿¡æ¯ï¼ˆå³ï¼šæ¯ç²’æ ·æœ¬æ‰€æœ‰åƒç´ å ‰è°±ä¿¡æ¯çš„å¹³å‡å€¼ï¼‰è¯†åˆ«æ•ˆæžœå¥½ï¼›åœ¨ç›¸åŒæƒ å†µä¸‹ï¼ŒCNN模型比SVM和KNNæ¨¡åž‹çš„è¯†åˆ«æ•ˆæžœå¥½ï¼›åŸºäºŽåƒç´ ‰è°±ä¿¡æ¯å’ŒCNNçš„å“ç§è¯†åˆ«æ¨¡åž‹è¯†åˆ«æ•ˆæžœæœ€ç¨³å®šï¼Œä¾æ®åƒç´ çº§åˆ†ç±»ç»“æžœé‡‡ç”¨å¤šæ•°æŠ•ç¥¨ç­–ç•¥å¯¹çŽ‰ç±³ç§å­æ ·æœ¬è¿›è¡Œè¯†åˆ«ï¼Œæ ·æœ¬è¯†åˆ«ç²¾åº¦é«˜è¾¾100%(注:100%ä¸ºå»ºæ¨¡é›†æ ·æœ¬ä¸Žæµ‹è¯•é›†æ ·æœ¬æ•°é‡ä¸º0.27和0.32æ—¶çš„è¯†åˆ«ç²¾åº¦ï¼Œéšç€æµ‹è¯•é›†æ ·æœ¬æ•°é‡çš„å¢žåŠ ï¼Œè¯¥è¯†åˆ«ç²¾åº¦å°†æœ‰æ‰€é™ä½Žï¼‰ã€‚æœ€åŽï¼Œä½¿ç”¨tåˆ†å¸ƒéšæœºé‚»åŸŸåµŒå ¥ï¼ˆt-SNE)算法实现CNN输出特征值的可视化,验证了基于高å 级信息和CNNçš„å“ç§è¯†åˆ«æ¨¡åž‹çš„æœ‰æ•ˆæ€§ã€‚åœ¨å»ºæ¨¡æ ·æœ¬æžå°‘çš„æƒ å†µä¸‹ï¼Œå®žçŽ°äº†çŽ‰ç±³ç§å­å“ç§çš„æ— æŸã€é«˜æ•ˆè¯†åˆ«ï¼Œä¸ºç²¾å‡†å†œä¸šæä¾›äº†ç†è®ºåŸºç¡€ã€‚

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Variety Identification of Single Rice Seed Using Hyperspectral Imaging Combined with Convolutional Neural Network

The feasibility of using hyperspectral imaging with convolutional neural network (CNN) to identify rice seed varieties was studied. Hyperspectral images of 4 rice seed varieties at two different spectral ranges (380–1030 nm and 874–1734 nm) were acquired. The spectral data at the ranges of 441–948 nm (Spectral range 1) and 975–1646 nm (Spectral range 2) were extracted. K nearest neighbors (KNN)...

متن کامل

Quad-pixel edge detection using neural network

One of the most fundamental features of digital image and the basic steps in image processing, analysis, pattern recognition and computer vision is the edge of an image where the preciseness and reliability of its results will affect directly on the comprehension machine system made objective world. Several edge detectors have been developed in the past decades, although no single edge detector...

متن کامل

Quad-pixel edge detection using neural network

One of the most fundamental features of digital image and the basic steps in image processing, analysis, pattern recognition and computer vision is the edge of an image where the preciseness and reliability of its results will affect directly on the comprehension machine system made objective world. Several edge detectors have been developed in the past decades, although no single edge detector...

متن کامل

Art Painting Identification using Convolutional Neural Network

Convolutional Neural Network (CNN) applications have been suggested for many multimedia processing tasks and achieved great success. In this paper, we present a methodology about how to apply CNN for art painting identification. Each art painting image is distorted by various operations, such as lens distortion, scaling, rotation, etc., to simulate potential situation that it would be appeared ...

متن کامل

Application of Hyperspectral Imaging and Chemometric Calibrations for Variety Discrimination of Maize Seeds

Hyperspectral imaging in the visible and near infrared (VIS-NIR) region was used to develop a novel method for discriminating different varieties of commodity maize seeds. Firstly, hyperspectral images of 330 samples of six varieties of maize seeds were acquired using a hyperspectral imaging system in the 380-1,030 nm wavelength range. Secondly, principal component analysis (PCA) and kernel pri...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Journal of remote sensing

سال: 2021

ISSN: ['1007-4619', '2095-9494']

DOI: https://doi.org/10.11834/jrs.20219349