Flower classification via convolutional neural network

Liu, Yuanyuan; Tang, Fan; Zhou, Dengwen; Ye, Meng; Dong, Weiming

doi:10.1109/fspma.2016.7818296

Cited by 53 publications

(23 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning techniques can be used to classify different flower types, which could be useful to determine the age of cotton flowers based on differences in color and shape (Seeland et al, 2016 ). Deep learning methods such as the convolutional neural network (CNN) have been demonstrated to be effective in recognizing flower species (Liu et al, 2016 ). CNN showed advantages over traditional machine learning methods because it does not require extraction of image features.…”

Section: Introductionmentioning

confidence: 99%

Aerial Images and Convolutional Neural Network for Cotton Bloom Detection

Paterson

et al. 2018

Front. Plant Sci.

View full text Add to dashboard Cite

Monitoring flower development can provide useful information for production management, estimating yield and selecting specific genotypes of crops. The main goal of this study was to develop a methodology to detect and count cotton flowers, or blooms, using color images acquired by an unmanned aerial system. The aerial images were collected from two test fields in 4 days. A convolutional neural network (CNN) was designed and trained to detect cotton blooms in raw images, and their 3D locations were calculated using the dense point cloud constructed from the aerial images with the structure from motion method. The quality of the dense point cloud was analyzed and plots with poor quality were excluded from data analysis. A constrained clustering algorithm was developed to register the same bloom detected from different images based on the 3D location of the bloom. The accuracy and incompleteness of the dense point cloud were analyzed because they affected the accuracy of the 3D location of the blooms and thus the accuracy of the bloom registration result. The constrained clustering algorithm was validated using simulated data, showing good efficiency and accuracy. The bloom count from the proposed method was comparable with the number counted manually with an error of −4 to 3 blooms for the field with a single plant per plot. However, more plots were underestimated in the field with multiple plants per plot due to hidden blooms that were not captured by the aerial images. The proposed methodology provides a high-throughput method to continuously monitor the flowering progress of cotton.

show abstract

Section: Introductionmentioning

confidence: 99%

Aerial Images and Convolutional Neural Network for Cotton Bloom Detection

Paterson

et al. 2018

Front. Plant Sci.

View full text Add to dashboard Cite

show abstract

“…Liu Yuanyuan [17] firstly established a larger dataset of 52,775 flower images in 79 categories. And a new model based on convolution neural network is proposed, which consists of five convolution layers, each convolution layer is followed by the largest pool layer, and then connects with three fully connected layers and the softmax layer.…”

Section: Methods Based On Deep Learningmentioning

confidence: 99%

A Flower Image Classification Algorithm Based on Saliency Map and PCANet

Yangyang¹,

Fu²

2019

JCC

View full text Add to dashboard Cite

Flower Image Classification is a Fine-Grained Classification problem. The main difficulty of Fine-Grained Classification is the large inter-class similarity and the inner-class difference. In this paper, we propose a new algorithm based on Saliency Map and PCANet to overcome the difficulty. This algorithm mainly consists of two parts: flower region selection, flower feature learning. In first part, we combine saliency map with gray-scale map to select flower region. In second part, we use the flower region as input to train the PCANet which is a simple deep learning network for learning flower feature automatically, then a 102-way softmax layer that follow the PCANet achieve classification. Our approach achieves 84.12% accuracy on Oxford 17 Flowers dataset. The results show that a combination of Saliency Map and simple deep learning network PCANet can applies to flower image classification problem.

show abstract

“…Convolutional Neural Networks distinguishes and differentiates objects or aspects from one another by assigning learnable weights or biases to various objects in the input image (Saha 2018). Liu et al (2016b) reported the same algorithm to be effective in identifying flower species. Although results from Xu et al (2018a) confirmed that the system developed for identifying and automatic counting of cotton flower was comparable with the results from manual counting, one disadvantage which was emphasized by the proponents was the underestimation in bloom counts when data were collected from a single plot with multiple crop stands.…”

Section: Other High-resolution Applicationsmentioning

confidence: 95%

High-throughput phenotyping in cotton: a review

et al. 2019

View full text Add to dashboard Cite

Recent technological advances in cotton (Gossypium hirsutum L.) phenotyping have offered tools to improve the efficiency of data collection and analysis. High-throughput phenotyping (HTP) is a non-destructive and rapid approach of monitoring and measuring multiple phenotypic traits related to the growth, yield, and adaptation to biotic or abiotic stress. Researchers have conducted extensive experiments on HTP and developed techniques including spectral, fluorescence, thermal, and three-dimensional imaging to measure the morphological, physiological, and pathological resistance traits of cotton. In addition, ground-based and aerial-based platforms were also developed to aid in the implementation of these HTP systems. This review paper highlights the techniques and recent developments for HTP in cotton, reviews the potential applications according to morphological and physiological traits of cotton, and compares the advantages and limitations of these HTP systems when used in cotton cropping systems. Overall, the use of HTP has generated many opportunities to accurately and efficiently measure and analyze diverse traits of cotton. However, because of its relative novelty, HTP has some limitations that constrains the ability to take full advantage of what it can offer. These challenges need to be addressed to increase the accuracy and utility of HTP, which can be done by integrating analytical techniques for big data and continuous advances in imaging.

show abstract

Flower classification via convolutional neural network

Cited by 53 publications

References 16 publications

Aerial Images and Convolutional Neural Network for Cotton Bloom Detection

Aerial Images and Convolutional Neural Network for Cotton Bloom Detection

A Flower Image Classification Algorithm Based on Saliency Map and PCANet

High-throughput phenotyping in cotton: a review

Contact Info

Product

Resources

About