Flower Images Research Articles

Jasminum Grandiflorum Flower Images Classification: Deep Learning and Transfer Learning Models with the Influence of Preprocessing via Contours and Convex Hull in Agritech 4.0

Jasminum Grandiflorum Flower Images Classification: Deep Learning and Transfer Learning Models with the Influence of Preprocessing via Contours and Convex Hull in Agritech 4.0

How do different categories of nature‐related landscape images shape emotional experience?

The aim of this study was to explore whether and how diverse categories of nature‐related landscape images would influence the emotional dimensions of pleasure, arousal and dominance in the process of aesthetic appreciation, as well as to develop a new nature‐related image system (NIS). A total of 575 high‐quality images across six categories were selected and assessed by 197 students from the dimensions of aesthetic feeling (AF), pleasure, arousal and dominance. The results showed that different categories evoked varying levels of each dimension, and the impact of landscape categories on emotional experiences cannot be overlooked. Specially, flower landscape showed a clear experimental dissociation in terms of AF and pleasure compared to other categories, suggesting that flower images tend to evoke a positive emotional experience overall, regardless of AF level. Moreover, it also suggested that the effects of flowers on emotions could be considered independently from other categories in future research.

Flower Recognition System using CNN

Flower recognition stands as a pivotal task across diverse fields such as botany, agriculture, and environmental studies. The manual process of identifying flower species is not only laborious but also susceptible to inaccuracies. In response to this challenge, this report introduces a Flower Recognition System employing Convolutional Neural Networks (CNNs) to automate and enhance this process. By harnessing the capabilities of deep learning, our system demonstrates remarkable proficiency in accurately classifying various flower species from input images. The CNN architecture, meticulously designed and trained on a comprehensive dataset of labeled flower images, exhibits robustness and efficiency in flower identification tasks. Through rigorous experimentation and evaluation, this report meticulously examines the development and performance of the Flower Recognition System. Furthermore, it explores the potential applications of this system in facilitating automated flower species identification, thereby contributing to advancements in botanical research, agricultural practices, and environmental conservation efforts.

Phase gradient metasurface for arbitrary three-dimensional shaped near-field

This paper proposes a methodology for generating arbitrary three-dimensional (3D) shaped near-field using phase gradient metasurface (PGMS) based on phase conjugation (PC) backtracking and shaped diffraction-free Bessel beam array techniques. The PGMS is designed by combining MATLAB and electromagnetic (EM) simulation software (CST). To the knowledge of the authors, for the first time, a 3D-shaped near-field is generated without increasing cost, structural and calculating complexity, and a methodology for generating arbitrarily shaped diffraction-free beam array is developed by MATLAB. To verify the technique, a planar PGMS is designed for dual x-polarized 3D-shaped near-fields: qq image “” in direction (θ1 = 5°, φ1 = 0°) and a flower image “” in direction (θ2 = 10°, φ2 = 180°). The advantages of the 3D-shaped near-fields are as follows: high conversion efficiency 30.5% (measured), wideband 60.5%, and low sidelobe -15 dB. The proposed technique is confirmed by the calculated, simulated, and measured results.

2024/2025년 F/W 팬톤 컬러 트렌드를 활용한 5종의 네일 믹스드 미디어의 개발

The colors that Pantone releases each year have a significant impact on many industries. Nail art is no exception to this trend, with many creative designs using seasonal colors. Previous studies have analyzed nail design works using Pantone colors every year, including 2018, 2020, and 2021, but there is a lack of nail mixed media research on Pantone color trends for 2024. The purpose of this study is to propose five nail mixed media based on the study of the characteristics of each of the five NYFW 2024/2025 New Classics colors and the results of image conception and analysis. To do so, we examined the New Classics colors, analyzed the characteristics and images of each of the five colors (White Swan, Raw Umber, Dark Gull Gray, Baritone Blue, and Buckthorn Brown), and developed nail mixed media pieces. As a result, first, the unique features of the New Classics colors, White Swan, Raw Umber, Dark Gull Gray, Baritone Blue, and Buckthorn Brown, were utilized and applied to the design. Second, the five color features and images were conceived and expressed as a swan image of White Swan, an elegant flower image of Raw Umber, a sophisticated image of Dark Gull Gray houndstooth, a mysterious image of the deep sea of Baritone Blue, and a natural image of Buckthorn Brown with an antique feel. These results show that Pantone colors are an influential part of nail design. They can be used to create new designs through various techniques that deviate from the usual nail art, It is recommended as a follow-up study to effectively utilize color trends, present differentiated styles, and develop multi-faceted nail trends.

Artificial intelligence-based smart agricultural systems for saffron cultivation with integration of Unmanned Aerial Vehicle imagery and deep learning approaches

This paper presents an integrated Unmanned Aerial Vehicle (UAV) imagery system and artificial intelligence-based smart farming for saffron cultivation. This work is about monitoring the growth of saffron flowers through UAV imagery every week or month during the pre-harvesting time while the percentage of the covered area by leaf and saffron flowers is calculated. This is crucial for predicting saffron growing areas over cultivated regions. From the vertical viewing of UAV imagery from above, it is problematic from the bird’s eye perspective to identify the particular saffron-growing areas of things, so the multi-class classification of flowers (subjects) recognition system is proposed. So, the implementation of this work is divided into four phases. The first two phases (Phase 1 and Phase 2) are about the algorithms for Saffron growing region detections and predictions using the modified You Only Look Once (YOLO) higher version model followed by statistical analysis based region corner detections. The outcomes of the first two phases are used to form a database of saffron flowers with different scales, orientations, and deformations. This is further used to build a two-class classification model in the third phase (Phase 3) that discriminates saffron vs. non-saffron flowers. In the fourth phase, the two-class classification model is extended to a multi-class flower recognition model for recognizing 900 flower images database available in the largest social media website dedicated exclusively to gardening in the Phase 4. Extensive experiments are conducted, and the performances are compared with some existing state-of-the-art methods that show the superiority of the proposed system.

Orchid2024: A cultivar-level dataset and methodology for fine-grained classification of Chinese Cymbidium Orchids

BackgroundChinese Cymbidium orchids, cherished for their deep-rooted cultural significance and significant economic value in China, have spawned a rich tapestry of cultivars. However, these orchid cultivars are facing challenges from insufficient cultivation practices and antiquated techniques, including cultivar misclassification, complex identification, and the proliferation of counterfeit products. Current commercial techniques and academic research primarily emphasize species identification of orchids, rather than delving into that of orchid cultivars within species.ResultsTo bridge this gap, the authors dedicated over a year to collecting a cultivar image dataset for Chinese Cymbidium orchids named Orchid2024. This dataset contains over 150,000 images spanning 1,275 different categories, involving visits to 20 cities across 12 provincial administrative regions in China to gather pertinent data. Subsequently, we introduced various visual parameter-efficient fine-tuning (PEFT) methods to expedite model development, achieving the highest top-1 accuracy of 86.14% and top-5 accuracy of 95.44%.ConclusionExperimental results demonstrate the complexity of the dataset while highlighting the considerable promise of PEFT methods within flower image classification. We believe that our work not only provides a practical tool for orchid researchers, growers and market participants, but also provides a unique and valuable resource for further exploring fine-grained image classification tasks. The dataset and code are available at https://github.com/pengyingshu/Orchid2024.

YNU Flower: A market-oriented benchmark dataset for flower classification

The great cultural values of flowers in our life have nourished a highly developed flower industry. However, our current flower trade is still labour-intensive mainly due to a lack of effective automated classification systems. While several image datasets have been established and are widely used for developing visual flower classification models, they are of little practical value to the industry as, in these datasets, few flowers of market value are included, and intra-class and temporal variabilities are also not taken into consideration. Here, we present a large-scale flower image dataset named YNU Flower for flower classification. A total of 91 commercial flower varieties belonging to eleven favoured botanical groups were purchased from the Dounan flower market, the largest flower trading centre in Asia, and 17,806 images were captured. Notably, the YNU Flower dataset includes many commercial varieties with subtle appearance distinctions but significantly different price tags. Using this newly established dataset, we trained several representative models for classification and object detection tasks. By making YNU Flower publicly available to the community, we anticipate that this benchmark dataset will expedite translational research on flower classification and ultimately benefit the flower industry.

SA-ConvNeXt: A Hybrid Approach for Flower Image Classification Using Selective Attention Mechanism

In response to the current lack of annotations for flower images and insufficient focus on key image features in traditional fine-grained flower image classification based on deep learning, this study proposes the SA-ConvNeXt flower image classification model. Initially, in the image preprocessing stage, a padding algorithm was used to prevent image deformation and loss of detail caused by scaling. Subsequently, the model was integrated using multi-level feature extraction within the Efficient Channel Attention (ECA) mechanism, forming an M-ECA structure to capture channel features at different levels; a pixel attention mechanism was also introduced to filter out irrelevant or noisy information in the images. Following this, a parameter-free attention module (SimAM) was introduced after deep convolution in the ConvNeXt Block to reweight the input features. SANet, which combines M-ECA and pixel attention mechanisms, was employed at the end of the module to further enhance the model’s dynamic extraction capability of channel and pixel features. Considering the model’s generalization capability, transfer learning was utilized to migrate the pretrained weights of ConvNeXt on the ImageNet dataset to the SA-ConvNeXt model. During training, the Focal Loss function and the Adam optimizer were used to address sample imbalance and reduce gradient fluctuations, thereby enhancing training stability. Finally, the Grad-CAM++ technique was used to generate heatmaps of classification predictions, facilitating the visualization of effective features and deepening the understanding of the model’s focus areas. Comparative experiments were conducted on the Oxford Flowers102 flower image dataset. Compared to existing flower image classification technologies, SA-ConvNeXt performed excellently, achieving a high accuracy of 96.7% and a recall rate of 98.2%, with improvements of 4.0% and 3.7%, respectively, compared to the original ConvNeXt. The results demonstrate that SA-ConvNeXt can effectively capture more accurate key features of flower images, providing an effective technical means for flower recognition and classification.

My favourite flowering image: the remarkable inside-out flowers of Lacandonia

Abstract Flowers are deceptively simple structures, characterized by a determinate primary axis that bears organs in condensed concentric zones in a strict structural and temporal sequence. Few species have escaped these constraints, but those that have can provide insights into the evolutionary history of flowers if placed in the appropriate phylogenetic and developmental context. For my flowering image, I selected a longitudinal section of a Lacandonia flower, which breaks a fundamental rule of spatial arrangement: the flowers are ‘inside-out’, with the carpels surrounding the stamens—a pattern that is almost unique among angiosperms. When viewed in the context of the family and order to which it belongs, this species has led me into many fascinating areas of comparative and evolutionary plant morphology.

Boea cavernarum Zich & B.Gray (Gesneriaceae), a remarkable new species from north Queensland, Australia

Boea cavernarum, restricted to near limestone crevices and caves in NE Queensland, is described. Images of whole plants, flowers, fruit and leaf indumentum, a distribution map, and a key to all species of Boea in Australia are provided. Thirteen species are now recognised in Boea.

Pre-training CNN model Several Methods of Comparative Analysis

In recent years, the rapid evolution of neural networks, exemplified by models such as VGG and GoogLeNet, has reshaped the landscape of artificial intelligence. This comparative exploration, focusing on transfer learning and attention mechanisms, not only reveals subtle performance distinctions but also provides crucial insights for refining these models in practical applications. This paper delves into the comparative analysis of two prominent neural network models, VGG and GoogLeNet, within the context of image classification. VGG, known for its deep structure, and GoogLeNet, incorporating innovative inception modules, are evaluated through experiments involving flower image datasets. The study includes transfer learning and explores the integration of attention mechanisms, specifically CBAM attention, into both models. Results indicate GoogLeNet's superior performance in terms of parameter efficiency, convergence speed, and overall accuracy. Furthermore, the addition of attention mechanisms enhances classification accuracy for both models. The paper concludes with insights into potential areas for further research, emphasizing optimal attention mechanism placement and a comparison with traditional methods in future studies.

Production of Hair Art Based on Rare Species of Flowers

Flowers are the outcome of plants. They have been commonly used as a material for art because they can express beauty with various colors and make diverse expressions possible. With the growth and development of the beauty art industry, people’s interest in hair art has increased. At the same time, materials for hair art have become diversified. For example, waste hair has been used for painterly-style art. In addition, color is emphasized for a formative meaning. In hair art created with flower images, hair is used to express flowers' diverse colors and 3-dimensional effects. The study results found the following: In Design I, dark colors were used to dynamically express the natural colors and vigor of wildflowers. In Design II, wildflowers were expressed like living creatures with the movements of petals. In Design III, dark colors were used to make flower stalks look like the janggu (Korean drum) stick, naturally expressing petals. In Design IV, the lively energy of the color yellow was expressed with flowers, vibrantly expressing leaves. In Design V, dark colors were used to create a straight look, expressing flowers and leaves with a natural violet color. In Design VI, the vigor of wildflowers was expressed with the diverse colors of flowers and petals, while their liveliness was created with leaves. This study further developed hair art into a practical field of study with rare flowers in the pictorial book of wildflowers, allowing anyone to make various attempts.

Further exploration on deep learning in flower recognition

Flower recognition is an important research direction in the field of computer vision, and automatic classification of flower images through deep learning methods is of great significance for ecological environment monitoring and plant research. With this background, this study aims to further optimize the existing flower recognition system and improve its classification accuracy by adjusting key parameters. Based on the existing deep learning model and several rounds of training, this paper explores the tuning strategies for parameters such as different learning rates and weight decay to achieve higher recognition accuracy. In the experiments, this paper uses the classical flower dataset and enhances the diversity of the data through image preprocessing and data enhancement. Through a hundred rounds of training, the model in this paper achieves about 80% classification accuracy on the test set, which is significantly improved compared with the initial model. Further analysis of the results shows that by reasonably adjusting the learning rate and weight decay parameters, the model achieves certain improvements in different flower classes, demonstrating the impact of parameter tuning on the overall performance of the model.

Revisiting the Watching Eyes Effect: How Emotional Expressions, Sex, and Age of Watching Eyes Influence Stereotypical Statement Endorsement

Abstract The watching eyes effect has been shown to influence prosocial and antisocial behaviors. However, the eye characteristics which induce this effect remain unclear. This study explored how emotional expressions (anger, fear, happy, neutral), age (old, young), and sex (male, female) of eye images affect antisocial behavior, measured by stereotype endorsement. Participants rated their endorsement of 36 stereotype statements about race, gender, and religion topics, each paired with an eye image. Our findings indicated that stereotype endorsement did not differ significantly between neutral eye images and control flower images. We then used neutral eyes as control images to evaluate stereotype endorsement when being watched by eyes with other expressions. When comparing endorsement across age and sex in each expression, the data revealed higher endorsement for angry old male eyes and for happy young eyes (males and females), and lower endorsement for fearful eyes, except for old fearful female eyes. Therefore, varying the emotional expression, sex, and age of the eye images used in the watching eyes effect paradigm influences stereotype endorsement. To further unravel the underlying mechanisms behind the watching eyes effect, we encourage future research to explore how varying characteristics of the eyes affect responses/behaviors like stereotype endorsement.

Comparison of the suitability of the otsu method thresholding and multilevel thresholding for flower image segmentation

The digital representation of flowers, characterized by their vivid chromatic attributes, establishes them as viable candidates for deployment as input imagery within the object recognition paradigm. Within the context of object recognition, the imperative of a proficient image segmentation process is underscored, serving to effectively discern the object from its background and, consequently, optimizing the efficacy of the object recognition process. This research unfolds through a methodologically structured tripartite framework, encompassing the initial stage involving input imagery, the subsequent intermediate phase dedicated to image segmentation, and a conclusive stage centered on the quantitative evaluation of methodological outcomes. The second stage, focusing on image segmentation, employs the Otsu thresholding and multilevel thresholding methods. The subsequent third stage involves a thorough assessment of segmentation outcomes through the application of quantitative metrics, including Peak signal-to-oise ratio (PSNR) and Root Mean Square Error (RMSE). Empirical investigations, incorporating a diverse array of floral input images, reveal a conspicuous inclination towards a specific segmentation methodology. Specifically, the Otsu Thresholding method emerges as the more judicious choice relative to multilevel Thresholding, demonstrating superior performance with a diminished RMSE value and an augmented PSNR value, substantiated by an average RMSE value. This research is propelled by the overarching objective of discerning the most optimal method for the segmentation of flower images, particularly in the face of diverse input images. Its significant contribution lies in providing nuanced insights into the discerning selection of segmentation methodologies, attuned to the variability inherent in diverse forms of input imagery, thereby culminating in optimized outcomes within the domain of flower image recognition. Where did these results come from? please show it in the sub-discussion.

Comparative Constructions with Images of Flowers in Modern Russian Prose

The subject of study are comparative constructions (metaphors and similes) which include the names of flowers. The material for the study is the texts of modern Russian prose, as well as contexts extracted from the Russian National Corpus. The purpose of the paper is to analyze these constructions in comparison with the tropes of the previous period. The images of comparison, the objects of comparison and the bases for comparison of comparative tropes have been considered, their semantic classification is presented. In the course of the study, the
 following conclusions were drawn: images of flowers are not as productive in modern prose as compared to other semantic classes of images of comparison, such as classes of zoonyms and names of artifacts; at the same time, they form a
 significant fragment of the metaphorical world image in modern prose. New elements appear in the composition of metaphors and similes with images of flowers, the range of objects of comparison is also expanding. In modern prose, traditional metaphorical models are preserved, but they are rethought and their character is partially changed. The use of flower images in modern texts is predominantly anthropocentric with an emphasis on the human bodily plane, which distinguishes them from functioning in the previous period. Modern prose is characterized by a tendency to lower the image when using the considered comparative tropes.
 Comparative constructions with images of flowers act in modern texts in various functions: a function of figurative characteristics of various objects, an evaluative function, and, more rarely, an intertextual one.

On generated artistic styles: Image generation experiments with GAN algorithms

As computer graphics technology supports pursuing a photorealistic style, replicated artworks with a photorealistic style overwhelmingly predominate in the computer-generated art circle. Along with the progression of generative technology, this trend may make generative art a virtual world of photorealistic fake, in which the single criterion of expressive style imperils art into the context of a single boring stereotype. This article focuses on the issue of style diversity and its technical feasibility by artistic experiments of generating flower images in StyleGAN. The author insisted that photo both technology and artistic style should not be confined merely for realistic purposes. This proposition was validated in the GAN generation experiment by changing the training materials.

Tomato flower pollination features recognition based on binocular gray value-deformation coupled template matching

It is necessary to recognize the tomato pollination features for the designing demand of intelligent and precise tomato supplementary pollination equipment. Mentioned pollination features include flower opening state and the three-dimensional position and pose of flower anther. Tomato flower pollination features recognition method is designed in this paper based on the deep learning full opened flower recognition model and binocular template matching three-dimensional information recognition method. First of all, a binocular stereo vision system is built to acquire the tomato flower images in greenhouse with natural lighting. Which can help vision system avoid the impact of inconsistent light intensity on image recognition. The acquired images were equalized with three groups parameters to labeled images. And the improved MC-AlexNet deep learning model is established to recognize the full opened tomato flowers in the left image of image pair acquired with binocular vision system. Then, the template is created with the full opened flower recognition result of deep learning model in left image based on gray value and deformation template matching method. And the template matching is established to recognize the corresponding full opened flowers in the right image of image pair. Finally, with the template matching result, the anther segmentation is conducted to calculate three-dimensional position and pose of anther. The experimental results show that the accuracy of full opened tomato flowers recognition model is 96.23%. The average position recognition deviation of anther is 5.94 mm. And the average anther pose recognition deviation in three plane is 6.24° for the images that anthers can be observed. And the average time consuming is about 143.18 ms per image pair. It can be concluded that the method of binocular template matching established in this paper can fulfill the demand of supplementary pollination equipment design, and the research result lays a foundation for designing and improvement of tomato supplementary pollination equipment in greenhouse.

A text image generation model based on deep learning

The traditional text-image confrontation model utilizes a convolutional network in the discriminator to extract image features, yet this fails to involve the spatial relationship between underlying objects, resulting in a poor-quality generated image. To remedy this, a capsule network is proposed to improve the model. The convolutional network in the discriminator is replaced with a capsule network, thereby improving the robustness of the images. Through experiments on the Oxford-102 and CUB datasets, it has been found that the new model can effectively improve the quality of generated text-image. The FID value of the generated flower image decreased by 14.49%, and the FID value of the generated bird image decreased by 9.64%. Additionally, the Inception Score of images generated on the Oxford-102 and CUB datasets increased by 22.60% and 26.28%, respectively, indicating that the improved model generated richer and more meaningful image features.