Deep learning for grape variety recognition

Franczyk, Bogdan; Hernes, Marcin; Kozierkiewicz, Adrianna; Kozina, Agata; Pietranik, Marcin; Roemer, Ingolf; Schieck, Martin

doi:10.1016/j.procs.2020.09.117

Cited by 40 publications

(24 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Each image on the dataset only contains one grape grain and the background is pure white (the background pixels are all in white color). Based on the influence of background processing on classification performance in the above discussion of literature [20], we believe that the clear background is also an important reason for such high accuracy. In summary, the method proposed in this study could obtain satisfactory performance for grape varieties identification with the images collected under natural conditions.…”

Section: Performance Comparison With Using Deep Network Directlymentioning

confidence: 92%

“…Bogdan et al cascaded the output of ResNet50 into a multi-layer perceptron, which can improve the classification accuracy, but this method needs to train two models. The evaluated dataset literature [20] is the same as ours, but the background was removed and each cluster of grapes was extracted separately for training and testing and an accuracy of 99% was obtained. However, in [22], when the background was not removed (the dataset is exactly the same as ours), the accuracy dropped by 26%, from 99% to 63%.…”

Section: Performance Comparison With Using Deep Network Directlymentioning

confidence: 99%

“…A well-known benchmark dataset, which provided the instances from five different grape varieties taken from the field, was used for training and testing on the developed model. The test results showed that the classification accuracy of the model for different grape varieties can reach 99% [20]. El-Mashharawi et al adopted a CNN for the identification of grape varieties.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fused Deep Features-Based Grape Varieties Identification Using Support Vector Machine

2021

View full text Add to dashboard Cite

Proper identification of different grape varieties by smart machinery is of great importance to modern agriculture production. In this paper, a fast and accurate identification method based on Canonical Correlation Analysis (CCA), which can fuse different deep features extracted from Convolutional Neural Network (CNN), plus Support Vector Machine (SVM) is proposed. In this research, based on an open dataset, three types of state-of-the-art CNNs, seven species of deep features, and a multi-class SVM classifier were studied. First, the images were resized to meet the input requirements of a CNN. Then, the deep features of the input images were extracted by a specific deep features layer of the CNN. Next, two kinds of deep features from different networks were fused by CCA to increase the effective classification feature information. Finally, a multi-class SVM classifier was trained with the fused features. When applied to an open dataset, the model outcome shows that the fused deep features with any combination can obtain better identification performance than by using a single type of deep feature. The fusion of fc6 (in AlexNet network) and Fc1000 (in ResNet50 network) deep features obtained the best identification performance. The average F1 Score of 96.9% was 8.7% higher compared to the best performance of a single deep feature, i.e., Fc1000 of ResNet101, which was 88.2%. Furthermore, the F1 Score of the proposed method is 2.7% higher than the best performance obtained by using a CNN directly. The experimental results show that the method proposed in this paper can achieve fast and accurate identification of grape varieties. Based on the proposed algorithm, the smart machinery in agriculture can take more targeted measures based on the different characteristics of different grape varieties for further improvement of the yield and quality of grape production.

show abstract

Section: Performance Comparison With Using Deep Network Directlymentioning

confidence: 92%

Section: Performance Comparison With Using Deep Network Directlymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fused Deep Features-Based Grape Varieties Identification Using Support Vector Machine

2021

View full text Add to dashboard Cite

show abstract

“…To evaluate the 3DeepM architecture with respect to other works, those that use deep learning techniques for grape classification have been selected. We found two common practices when using deep learning methods for classification: (1) use predefined architectures [2,56] (see Section 2.3.1) and (2) develop ad hoc architectures for a specific problem [4,57,58].…”

Section: Grape Classificationmentioning

confidence: 99%

“…In [2] the AlexNet architecture, transfer learning is used for the identification of six grape bunch varieties with an accuracy of 77.30%. In [56], the authors present a classification model (ExtResnet) based on the extension of the Resnet architecture. The proposed architecture incorporates a block of FC layers to Resnet, together with a multiple-branch output.…”

Section: Grape Classificationmentioning

confidence: 99%

3DeepM: An Ad Hoc Architecture Based on Deep Learning Methods for Multispectral Image Classification

et al. 2021

View full text Add to dashboard Cite

Current predefined architectures for deep learning are computationally very heavy and use tens of millions of parameters. Thus, computational costs may be prohibitive for many experimental or technological setups. We developed an ad hoc architecture for the classification of multispectral images using deep learning techniques. The architecture, called 3DeepM, is composed of 3D filter banks especially designed for the extraction of spatial-spectral features in multichannel images. The new architecture has been tested on a sample of 12210 multispectral images of seedless table grape varieties: Autumn Royal, Crimson Seedless, Itum4, Itum5 and Itum9. 3DeepM was able to classify 100% of the images and obtained the best overall results in terms of accuracy, number of classes, number of parameters and training time compared to similar work. In addition, this paper presents a flexible and reconfigurable computer vision system designed for the acquisition of multispectral images in the range of 400 nm to 1000 nm. The vision system enabled the creation of the first dataset consisting of 12210 37-channel multispectral images (12 VIS + 25 IR) of five seedless table grape varieties that have been used to validate the 3DeepM architecture. Compared to predefined classification architectures such as AlexNet, ResNet or ad hoc architectures with a very high number of parameters, 3DeepM shows the best classification performance despite using 130-fold fewer parameters than the architecture to which it was compared. 3DeepM can be used in a multitude of applications that use multispectral images, such as remote sensing or medical diagnosis. In addition, the small number of parameters of 3DeepM make it ideal for application in online classification systems aboard autonomous robots or unmanned vehicles.

show abstract

Assessing farmers' perception to Agriculture 4.0 technologies: A new interval‐valued spherical fuzzy sets based approach

Erdoğan

2021

Int J of Intelligent Sys

View full text Add to dashboard Cite

Agriculture 4.0 is the usage of advanced technologies for ensuring sustainability and efficient use of resources in agriculture, which can be accepted one of the biggest challenges of mankind. With Agriculture 4.0 technologies, it is aimed to raise productivity, reduce waste and costs. To ensure that agriculture 4.0 technologies are adopted by farmers as soon as possible, the perspective and perception of farmers to these technologies should be analyzed firstly. For this purpose, we propose a decision-making framework to measure farmers' view of Agriculture 4.0 technologies and to perform a prioritization study based on the perception of use among these technologies. Multicriteria decision analysis is also utilized to deal with all qualitative and quantitative factors in the decision process. Within the scope of this study, interval-valued spherical fuzzy numbers are used to model the vagueness in the process in the best manner and to be able to reflect the uncertainty arising from the usage of linguistic variables in the decision process. The SWARA and MAIRCA multicriteria decision making (MCDM) methods, which have been used frequently in the literature and are applied very successfully in MCDM problems, have been firstly extended by spherical fuzzy sets (SFSs) and the advantages for these methods have been utilized within the framework of

show abstract

Deep learning for grape variety recognition

Cited by 40 publications

References 20 publications

Fused Deep Features-Based Grape Varieties Identification Using Support Vector Machine

Fused Deep Features-Based Grape Varieties Identification Using Support Vector Machine

3DeepM: An Ad Hoc Architecture Based on Deep Learning Methods for Multispectral Image Classification

Assessing farmers' perception to Agriculture 4.0 technologies: A new interval‐valued spherical fuzzy sets based approach

Contact Info

Product

Resources

About