Varietal Classification of Rice Seeds Using RGB and Hyperspectral Images

Abstract

Inspection of rice seeds is a crucial task for plant nurseries and farmers since it ensures seed quality when growing seedlings. Conventionally, this process is performed by expert inspectors who manually screen large samples of rice seeds to identify their species and assess the cleanness of the batch. In the quest to automate the screening process through machine vision, a variety of approaches utilise appearance-based features extracted from RGB images while others utilise the spectral information acquired using Hyperspectral Imaging (HSI) systems. Most of the literature on this topic benchmarks the performance of new discrimination models using only a small number of species. Hence, it is unclear whether or not model performance variance confirms the effectiveness of proposed algorithms and features, or if it can be simply attributed to the inter-class/intra-class variations of the dataset itself. In this paper, a novel method to automatically screen and classify rice seed samples is proposed using a combination of spatial and spectral features, extracted from high resolution RGB and hyperspectral images. The proposed system is evaluated using a large dataset of 8,640 rice seeds sampled from a variety of 90 different species. The dataset is made publicly available to facilitate robust comparison and benchmarking of other existing and newly proposed techniques going forward. The proposed algorithm is evaluated on this large dataset and the experimental results show the effectiveness of the algorithm to eliminate impure species by combining spatial features extracted from high spatial resolution images and spectral features from hyperspectral data cubes. INDEX TERMS Hyperspectral imaging, rice seed variety, spatio-temporal feature fusion. JINCHANG REN (Senior Member, IEEE) received the B.E. degree in computer software, the M.Eng. degree in image processing, and the D.Eng. degree in computer vision from Northwestern Polytechnical University, Xi'an, China, and the Ph.D. degree in electronic imaging and media communication from the

Big Data and Machine Learning With Hyperspectral Information in Agriculture

“…Predict biomass from LiDAR point clouds and hyperspectral data Hyperspectral remote sensing image classification [32] Support vector machine (SVM) Self-training method with spatial majority filtering to find unlabeled samples for SVM classifier training Vine water status prediction [33] Multilayer perceptron (MLP) to predict relation between spectral bands and vine water status Their results showed that the SVM produced the highest classification performance of 95.9% with the ground surveyed crop areas. The authors in [25] proposed a system for the classification of rice seed varieties using RGB and hyperspectral images. The spatial and spectral features were extracted from the RGB images and hyperspectral image data cubes.…”

Research on forage hyperspectral image recognition based on F-SVD and XGBoost

“…The spectrum ranges from 400 to 1000nm, 750 channels, line scanning, and including 125 bands, that images resolution is 1166×1004 pixels. It is necessary to perform radiation correction every ten minutes to avoid the influence of external sun, wind and other factors [10]. Fig.…”

PhosopNet: An improved grain localization and classification by image augmentation

“…From the literature, all papers could be categorized by methods [8] into 2 groups: bag of words [9][10][11][12][13][14][15][16][17][18][19][20][21][22][23] and convolutional neural network (CNN) [24][25][26][27][28][29]. The former was used by early researches [30] (which require less number of labeled data [31]) that were still useful in some open-world industry [20][21][22] such as iRSVPred [23]. The latter was exponentially increased by current researches (which required large volume of labeled data [32,33]) that had already been proven to be higher performance than bag-of-words methods [34].…”