- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Image based leaf segmentation and counting in rosette plants
摘要： This paper proposes an efficient method to extract the leaf region and count the number of leaves in digital plant images. The plant image analysis plays a significant role in viable and productive agriculture. It is used to record the plant growth, plant yield, chlorophyll fluorescence, plant width and tallness, leaf area, etc. frequently and accurately. Plant growth is a major character to be analyzed among these plant characters and it directly depends on the number of leaves in the plants. In this paper, a new method is presented for leaf region extraction from plant images and counting the number of leaves. The proposed method has three steps. The first step involves a new statistical based technique for image enhancement. The second step involves in the extraction of leaf region in plant image using a graph based method. The third step involves in counting the number of leaves in the plant image by applying Circular Hough Transform (CHT). The proposed work has been experimented on benchmark datasets of Leaf Segmentation Challenge (LSC). The proposed method achieves the segmentation accuracy of 95.4% and it also achieves the counting accuracy of (cid:1)0.7 (DiC) and 2.3 (|DiC|) for datasets (A1, A2 and A3), which are better than the state-of-the-art methods.
关键词： Leaf count,Plant image analysis,Plant phenotyping,Leaf region extraction
更新于2025-09-23 15:23:52
-
Optimized angles of the swing hyperspectral imaging system for single corn plant
摘要： During recent years, hyperspectral imaging systems have been widely applied in the greenhouses for plant phenotyping purposes. Current systems are typically designed as either top view or side view imaging mode. Top view is an ideal imaging angle for top leaves with flat leaf surfaces. However, most bottom leaves are either blocked or shaded. From side view, the entire plant structure is viewable. However, most leaf surfaces are not facing the camera, which impacts measurement quality. Besides, there could be advantages with certain tilted angle(s) between top view and side view. It’s interesting to explore the impact of different imaging angles to the phenotyping quality. For this purpose, a swing hyperspectral imaging system capable of capturing images at any angle from side view (0°) to top view (90°) by rotating the camera and the lighting source was designed. Corn plants were grown and allocated into 3 different treatments: high nitrogen (N) and well-watered (control), high N and drought-stressed, and low N and well-watered. Each plant was imaged at 7 different angles from 0° to 90° with an interval of 15°. The soil plant analysis development (SPAD) values and relative water content (RWC) ground truth measurements were used to establish treatment effects. The results showed that averaged plant-level Normalized Difference Vegetation Index (NDVI) values of plants in different treatments changed at different imaging angles. The results also indicated that for pixel-level NDVI distributions, the titled imaging angle of 75° was optimal to distinguish different water treatments, whereas, the tilted imaging angle of 15° was optimal to distinguish different N treatments. For pixel-level RWC distributions, the distribution difference between different water treatments was larger at higher imaging angles.
关键词： Pixel-level NDVI and RWC distributions,Optimal imaging angle,Swing hyperspectral imaging system,Plant phenotyping system,Tilted imaging angle
更新于2025-09-23 15:22:29
-
Spatial Referencing of Hyperspectral Images for Tracing of Plant Disease Symptoms
摘要： The characterization of plant disease symptoms by hyperspectral imaging is often limited by the missing ability to investigate early, still invisible states. Automatically tracing the symptom position on the leaf back in time could be a promising approach to overcome this limitation. Therefore we present a method to spatially reference time series of close range hyperspectral images. Based on reference points, a robust method is presented to derive a suitable transformation model for each observation within a time series experiment. A non-linear 2D polynomial transformation model has been selected to cope with the specific structure and growth processes of wheat leaves. The potential of the method is outlined by an improved labeling procedure for very early symptoms and by extracting spectral characteristics of single symptoms represented by Vegetation Indices over time. The characteristics are extracted for brown rust and septoria tritici blotch on wheat, based on time series observations using a VISNIR (400–1000 nm) hyperspectral camera.
关键词： spectral tracking,time series,plant phenotyping,hyperspectral imaging,disease detection
更新于2025-09-23 15:22:29
-
[IEEE 2018 IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI) - Las Vegas, NV (2018.4.8-2018.4.10)] 2018 IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI) - Estimating Plant Centers Using A Deep Binary Classifier
摘要： Phenotyping is the process of estimating the physical and chemical properties of a plant. Traditional phenotyping is labor intensive and time consuming. These measurements can be obtained faster by collecting aerial images with an Unmanned Aerial Vehicle (UAV) and analyzing them using modern image analysis technologies. We propose a method to estimate plant centers by classifying each pixel as a plant center or not a plant center. We then label the center of each cluster as the plant location. We studied the performance of our method on two datasets. We achieved 84% precision and 90% recall on one dataset consisting of early stage plants and 62% precision and 77% recall on another dataset consisting of later stage plants.
关键词： Color Image Processing,Plant Phenotyping,CNN,Machine Learning
更新于2025-09-23 15:21:21
-
A Cost-Effective and Portable Optical Sensor System to Estimate Leaf Nitrogen and Water Contents in Crops
摘要： Non‐invasive determination of leaf nitrogen (N) and water contents is essential for ensuring the healthy growth of the plants. However, most of the existing methods to measure them are expensive. In this paper, a low‐cost, portable multispectral sensor system is proposed to determine N and water contents in the leaves, non‐invasively. Four different species of plants—canola, corn, soybean, and wheat—are used as test plants to investigate the utility of the proposed device. The sensor system comprises two multispectral sensors, visible (VIS) and near‐infrared (NIR), detecting reflectance at 12 wavelengths (six from each sensor). Two separate experiments were performed in a controlled greenhouse environment, including N and water experiments. Spectral data were collected from 307 leaves (121 for N and 186 for water experiment), and the rational quadratic Gaussian process regression (GPR) algorithm was applied to correlate the reflectance data with actual N and water content. By performing five‐fold cross‐validation, the N estimation showed a coefficient of determination (??2) of 63.91% for canola, 80.05% for corn, 82.29% for soybean, and 63.21% for wheat. For water content estimation, canola showed an ??2 of 18.02%, corn showed an ??2 of 68.41%, soybean showed an ??2 of 46.38%, and wheat showed an ??2 of 64.58%. The result reveals that the proposed low‐cost sensor with an appropriate regression model can be used to determine N content. However, further investigation is needed to improve the water estimation results using the proposed device.
关键词： plant phenotyping,non‐invasive,machine learning,reflectance,leaf nitrogen
更新于2025-09-19 17:13:59
-
Ensemble Feature Selection for Plant Phenotyping: A Journey From Hyperspectral to Multispectral Imaging
摘要： Hyperspectral imaging is becoming an increasingly popular tool for high-throughput plant phenotyping, because it provides remarkable insights about the health status of plants. Feature selection is a key component in a hyperspectral image analysis, largely because a significant portion of spectral features are redundant and/or irrelevant, depending on the desired application. This paper presents an ensemble feature selection method to identify the most informative spectral features for practical applications in plant phenotyping. The hyperspectral data set contained the images of four wheat lines, each with a control and a salt (NaCl) treatment. To rank spectral features, six feature selection methods were used as the base for the ensemble: correlation-based feature selection, ReliefF, sequential feature selection, support vector machine-recursive feature elimination (SVM-RFE), LASSO logistic regression, and random forest. The best results were achieved by the ensemble of ReliefF, SVM-RFE, and random forest, which drastically reduced the dimension of the hyperspectral data set from 215 to 15 features, while improving the accuracy in classifying the salt-treated vegetation pixels from the control pixels by 8.5%. To transform the hyperspectral data set into a multispectral data set, six wavelengths as the center of broad multispectral bands around the most prominent features were determined by a clustering algorithm. The result of salt tolerance assessment of the four wheat lines using the derived multispectral data set was similar to that of the hyperspectral data set. This demonstrates that the proposed feature selection pipeline can be utilized for determining the most informative features and can be a valuable tool in the development of tailored multispectral cameras.
关键词： hyperspectral imaging,Band selection,multispectral imaging,wheat,ensemble feature selection,salt stress,machine learning,plant phenotyping,classification
更新于2025-09-10 09:29:36
-
[American Society of Agricultural and Biological Engineers 2017 Spokane, Washington July 16 - July 19, 2017 - ()] 2017 Spokane, Washington July 16 - July 19, 2017 - <i>Phenotyping of Arabidopsis for drought stress response using kinetic chlorophyll fluorescence imaging</i>
摘要： Drought stress is one of the major concerns in global agricultural production. Developing an efficient phenotyping technology can bridge the knowledge gap between the plant phenotype and genotype, which can promote the progress of breeding for drought tolerant accessions and provide economic benefits for the producers and consumers. This research was aimed to investigate the plant phenotyping for drought stress responses of two different genotypes of Arabidopsis using chlorophyll fluorescence imaging. 59 treatment groups (three plants for each group) of each genotype were withholding being watered for 8 days as the drought stress treatment, and the other 59 groups considered as control were regularly watered with 6 ml 1% nutrient solution every day. The kinetic chlorophyll fluorescence images of the drought treatment groups and the control groups were acquired at day 1, 3, 5, 7 and 8 after the drought stress treatment started. The conventional chlorophyll fluorescence parameters and the leaf area index were then extracted from the images. In addition, associated morphological and physiological parameters were also assayed. To construct combinatorial images, the sequential forward selection (SFS) algorithm was used to select the maximum contrast images between two genotypes and the linear discriminant analysis (LDA) was used to build combinatorial images. Finally, combinatorial images were analyzed, indicating combinatorial images are valuable in drought stress studies. Above all, the study showed that AQ and osca1 presented different drought stress responses during the treatment period based on the conventional chlorophyll parameters and combinatorial images.
关键词： drought stress,Arabidopsis,plant phenotyping,combinatorial imaging,Chlorophyll fluorescence imaging
更新于2025-09-10 09:29:36
-
Leaf Area Estimation of Reconstructed Maize Plants Using a Time-of-Flight Camera Based on Different Scan Directions
摘要： The leaf area is an important plant parameter for plant status and crop yield. In this paper, a low-cost time-of-flight camera, the Kinect v2, was mounted on a robotic platform to acquire 3-D data of maize plants in a greenhouse. The robotic platform drove through the maize rows and acquired 3-D images that were later registered and stitched. Three different maize row reconstruction approaches were compared: reconstruct a crop row by merging point clouds generated from both sides of the row in both directions, merging point clouds scanned just from one side, and merging point clouds scanned from opposite directions of the row. The resulted point cloud was subsampled and rasterized, the normals were computed and re-oriented with a Fast Marching algorithm. The Poisson surface reconstruction was applied to the point cloud, and new vertices and faces generated by the algorithm were removed. The results showed that the approach of aligning and merging four point clouds per row and two point clouds scanned from the same side generated very similar average mean absolute percentage error of 8.8% and 7.8%, respectively. The worst error resulted from the two point clouds scanned from both sides in opposite directions with 32.3%.
关键词： crop characterization,precision farming,3-D sensors,agricultural robotics,plant phenotyping
更新于2025-09-09 09:28:46
-
On-the-go hyperspectral imaging for the in-field estimation of grape berry soluble solids and anthocyanin concentration
摘要： Background and Aims: Hyperspectral imaging (HSI) is used to assess fruit composition mostly indoor under controlled conditions. This work evaluates a HSI technique to measure TSS and anthocyanin concentration in wine grapes non-destructively, in real time and in the vineyard. Methods and Results: Hyperspectral images were acquired under natural illumination with a VIS–NIR hyperspectral camera (400–1000 nm) mounted on an all-terrain vehicle moving at 5 km/h in a commercial Tempranillo vineyard in La Rioja, Spain. Measurements were taken on four dates during grape ripening in 2017. Grape composition was analysed on the grapes imaged, which was then used to develop spectral models, trained with support vector machines, to predict TSS and anthocyanin concentration. Regression models of TSS had determination coefficients (R2) of 0.91 for a fivefold cross validation [root mean squared error (RMSE) of 1.358°Brix] and 0.92 for the prediction of external samples (RMSE of 1.274°Brix). For anthocyanin concentration, R2 of 0.72 for cross validation (RMSE of 0.282 mg/g berry) and 0.83 for prediction (RMSE of 0.211 mg/g berry) was achieved. Spatial–temporal variation maps were developed for the four image acquisition dates during ripening. Conclusions: These results suggest that potential for on-the-go HSI to automate the assessment of important grape compositional parameters in vineyard is promising. Significance of the Study: The on-the-go HSI method described in this study could be automated and provide valuable information to improve winery and vineyard decisions and vineyard management.
关键词： sensors,plant phenotyping,support vector machines,proximal sensing regression
更新于2025-09-09 09:28:46