Plant Counts Research Articles

Biodiversity Assessment of Trees and Shrubs at Tirunelveli’s Science Centre in Southern India: A Conservation Perspective

A comprehensive taxonomical floristic survey conducted at the District Science Centre in Tirunelveli identified 84 species of trees and shrubs, documenting each plant species with detailed information on medicinal uses, family classification, commercial importance, and population count of individual plants. The survey identified 33 families and 68 genera, with Fabaceae being the most dominant family, followed by Rubiaceae, Arecaceae, and Malvaceae. Additionally, the survey highlighted a list of vulnerable, threatened, and endangered tree species such as Cycas beddomei Dyer., Santalum album L., emphasizing the need for conservation efforts based on their spatial reach and anthropogenic pressures. These species face various threats, including loss of habitat, agriculture, pollution load, and recreational disturbances, and were assessed according to the IUCN Red List criteria. This study provides valuable insights for botanists, conservationists, and policymakers, promoting sustainable management and protection of plant biodiversity in the region.

A Coffee Plant Counting Method Based on Dual-Channel NMS and YOLOv9 Leveraging UAV Multispectral Imaging

Accurate coffee plant counting is a crucial metric for yield estimation and a key component of precision agriculture. While multispectral UAV technology provides more accurate crop growth data, the varying spectral characteristics of coffee plants across different phenological stages complicate automatic plant counting. This study compared the performance of mainstream YOLO models for coffee detection and segmentation, identifying YOLOv9 as the best-performing model, with it achieving high precision in both detection (P = 89.3%, mAP50 = 94.6%) and segmentation performance (P = 88.9%, mAP50 = 94.8%). Furthermore, we studied various spectral combinations from UAV data and found that RGB was most effective during the flowering stage, while RGN (Red, Green, Near-infrared) was more suitable for non-flowering periods. Based on these findings, we proposed an innovative dual-channel non-maximum suppression method (dual-channel NMS), which merges YOLOv9 detection results from both RGB and RGN data, leveraging the strengths of each spectral combination to enhance detection accuracy and achieving a final counting accuracy of 98.4%. This study highlights the importance of integrating UAV multispectral technology with deep learning for coffee detection and offers new insights for the implementation of precision agriculture.

Integrating Automated Labeling Framework for Enhancing Deep Learning Models to Count Corn Plants Using UAS Imagery.

Plant counting is a critical aspect of crop management, providing farmers with valuable insights into seed germination success and within-field variation in crop population density, both of which are key indicators of crop yield and quality. Recent advancements in Unmanned Aerial System (UAS) technology, coupled with deep learning techniques, have facilitated the development of automated plant counting methods. Various computer vision models based on UAS images are available for detecting and classifying crop plants. However, their accuracy relies largely on the availability of substantial manually labeled training datasets. The objective of this study was to develop a robust corn counting model by developing and integrating an automatic image annotation framework. This study used high-spatial-resolution images collected with a DJI Mavic Pro 2 at the V2-V4 growth stage of corn plants from a field in Wooster, Ohio. The automated image annotation process involved extracting corn rows and applying image enhancement techniques to automatically annotate images as either corn or non-corn, resulting in 80% accuracy in identifying corn plants. The accuracy of corn stand identification was further improved by training four deep learning (DL) models, including InceptionV3, VGG16, VGG19, and Vision Transformer (ViT), with annotated images across various datasets. Notably, VGG16 outperformed the other three models, achieving an F1 score of 0.955. When the corn counts were compared to ground truth data across five test regions, VGG achieved an R2 of 0.94 and an RMSE of 9.95. The integration of an automated image annotation process into the training of the DL models provided notable benefits in terms of model scaling and consistency. The developed framework can efficiently manage large-scale data generation, streamlining the process for the rapid development and deployment of corn counting DL models.

LettuceNet: A Novel Deep Learning Approach for Efficient Lettuce Localization and Counting

Traditional lettuce counting relies heavily on manual labor, which is laborious and time-consuming. In this study, a simple and efficient method for localization and counting lettuce is proposed, based only on lettuce field images acquired by an unmanned aerial vehicle (UAV) equipped with an RGB camera. In this method, a new lettuce counting model based on the weak supervised deep learning (DL) approach is developed, called LettuceNet. The LettuceNet network adopts a more lightweight design that relies only on point-level labeled images to train and accurately predict the number and location information of high-density lettuce (i.e., clusters of lettuce with small planting spacing, high leaf overlap, and unclear boundaries between adjacent plants). The proposed LettuceNet is thoroughly assessed in terms of localization and counting accuracy, model efficiency, and generalizability using the Shanghai Academy of Agricultural Sciences-Lettuce (SAAS-L) and the Global Wheat Head Detection (GWHD) datasets. The results demonstrate that LettuceNet achieves superior counting accuracy, localization, and efficiency when employing the enhanced MobileNetV2 as the backbone network. Specifically, the counting accuracy metrics, including mean absolute error (MAE), root mean square error (RMSE), normalized root mean square error (nRMSE), and coefficient of determination (R2), reach 2.4486, 4.0247, 0.0276, and 0.9933, respectively, and the F-Score for localization accuracy is an impressive 0.9791. Moreover, the LettuceNet is compared with other existing widely used plant counting methods including Multi-Column Convolutional Neural Network (MCNN), Dilated Convolutional Neural Networks (CSRNets), Scale Aggregation Network (SANet), TasselNet Version 2 (TasselNetV2), and Focal Inverse Distance Transform Maps (FIDTM). The results indicate that our proposed LettuceNet performs the best among all evaluated merits, with 13.27% higher R2 and 72.83% lower nRMSE compared to the second most accurate SANet in terms of counting accuracy. In summary, the proposed LettuceNet has demonstrated great performance in the tasks of localization and counting of high-density lettuce, showing great potential for field application.

Canopy closure, yield, and quality under heterogeneous plant spacing in sunflower

AbstractDeviation from uniform target plant population density in sunflower (Helianthus annuus L.) production may negatively affect canopy closure, yield, and grain quality. The objective of this study was to evaluate the effects of plant density heterogeneity on yield, quality, and canopy closure in confectionery and oilseed sunflowers. Field experiments consisted of fixed differences in the number of skips/doubles per plot while maintaining equal total number of plants, and stand gaps with proportional decreasing plant counts. These were conducted in Minnesota, Texas, and Colorado, USA. Experiments revealed that yield was occasionally influenced by heterogeneity, with more uniform stand density yielding higher; however, substantial changes in uniformity of plant density often resulted in no differences in yield. Confectionery sunflowers compensated for yield losses in thinner stands mainly by producing larger seeds (observed range of 0.1%–46.2% of seeds over a 9.53 mm sieve), while oilseed sunflowers increased seed production per head (as shown by invariable test weight across treatments within environments). Plants that were unharvestable for any cause (most often due to head rot disease or lodging) were more prevalent in less uniform stands, and faster canopy closure was achieved in uniformly spaced plants. Although sunflower is very resilient to stand uniformity issues, these findings underscore the importance of uniform sunflower plant spacing for effective weed suppression and crop performance and highlight the need for further research into sunflower yield compensation mechanisms.

The value of generalized linear mixed models for data analysis in the plant sciences

Modern data analysis typically involves the fitting of a statistical model to data, which includes estimating the model parameters and their precision (standard errors) and testing hypotheses based on the parameter estimates. Linear mixed models (LMMs) fitted through likelihood methods have been the foundation for data analysis for well over a quarter of a century. These models allow the researcher to simultaneously consider fixed (e.g., treatment) and random (e.g., block and location) effects on the response variables and account for the correlation of observations, when it is assumed that the response variable has a normal distribution. Analysis of variance (ANOVA), which was developed about a century ago, can be considered a special case of the use of an LMM. A wide diversity of experimental and treatment designs, as well as correlations of the response variable, can be handled using these types of models. Many response variables are not normally distributed, of course, such as discrete variables that may or may not be expressed as a percentage (e.g., counts of insects or diseased plants) and continuous variables with asymmetrical distributions (e.g., survival time). As expansions of LMMs, generalized linear mixed models (GLMMs) can be used to analyze the data arising from several non-normal statistical distributions, including the discrete binomial, Poisson, and negative binomial, as well as the continuous gamma and beta. A GLMM allows the data analyst to better match the model to the data rather than to force the data to match a specific model. The increase in computer memory and processing speed, together with the development of user-friendly software and the progress in statistical theory and methodology, has made it practical for non-statisticians to use GLMMs since the late 2000s. The switch from LMMs to GLMMs is deceptive, however, as there are several major issues that must be thought about or judged when using a GLMM, which are mostly resolved for routine analyses with LMMs. These include the consideration of conditional versus marginal distributions and means, overdispersion (for discrete data), the model-fitting method [e.g., maximum likelihood (integral approximation), restricted pseudo-likelihood, and quasi-likelihood], and the choice of link function to relate the mean to the fixed and random effects. The issues are explained conceptually with different model formulations and subsequently with an example involving the percentage of diseased plants in a field study with wheat, as well as with simulated data, starting with a LMM and transitioning to a GLMM. A brief synopsis of the published GLMM-based analyses in the plant agricultural literature is presented to give readers a sense of the range of applications of this approach to data analysis.

Electric field-based air nanobubbles (EF-ANBs) irrigation on efficient crop cultivation with reduced fertilizer dependency

The advent of air nanobubbles (ANBs) has opened up a wide range of commercial applications spanning industries including wastewater treatment, food processing, biomedical engineering, and agriculture. The implementation of electric field-based air nanobubbles (EF-ANBs) irrigation presents a promising approach to enhance agricultural crop efficiency, concurrently promoting environmentally sustainable practices through reducing fertilizer usage. This study investigated the impact of EF-ANBs on the germination and overall growth of agricultural crops in soil. Results indicate a substantial enhancement in both germination rates and plant growth upon the application of EF-ANBs. Notably, the introduction of ANBs led to a significant enhancement in the germination rate of lettuce and basil, increasing from approximately 20% to 96% and from 16% to 53%, respectively over two days. Moreover, the presence of EF-ANBs facilitates superior hypocotyl elongation, exhibiting a 2.8- and a 1.6-fold increase in the elongation of lettuce and basil, respectively, over a six-day observation period. The enriched oxygen levels within the air nanobubbles expedite aerobic respiration, amplifying electron leakage from the electron transport chain (ETC) and resulting in heightened reactive oxygen species (ROS) production, playing a pivotal role in stimulating growth signaling. Furthermore, the application of EF-ANBs in irrigation surpasses the impact of traditional fertilizers, demonstrating a robust catalytic effect on the shoot, stem, and root length, as well as the leaf count of lettuce plants. Considering these parameters, a single fertilizer treatment (at various concentrations) during EF-ANBs administration, demonstrates superior plant growth compared to regular water combined with fertilizer. The findings underscore the synergistic interaction between aerobic respiration and the generation of ROS in promoting plant growth, particularly in the context of reduced fertilizer levels facilitated by the presence of EF-ANBs. This promising correlation holds significant potential in establishing more sustainability for ever-increasing environmentally conscious agriculture.

Effectiveness of integrating mowing and systemic herbicides applied with a weed wiper for Sporobolus indicus var. pyramidalis management in Florida

Abstract Giant smutgrass [Sporobolus indicus (L.) R. Br. var. pyramidalis (P. Beauv.) Veldkamp] is an invasive species in grasslands, and herbicide application has been the most efficient management method to suppress this weed. Experiments were conducted in 2017 and 2018 to determine the effects of wiping glyphosate and hexazinone on S. indicus var. pyramidalis. A dose–response experiment using a handheld weed wiper was established with 20 treatments comprising two herbicides (glyphosate and hexazinone), uni- and bidirectional wiping methods, and 5 herbicide concentrations (6.25% v/v, 12.5% v/v, 25.0% v/v, 50.0% v/v, and 100% v/v basis). Data were collected 30 and 60 d after treatment (DAT). An ATV-mounted roto-type weed-wiper experiment was established in a strip-plot arrangement, with mowing as the horizontal strip, the wiping method (unidirectional vs. bidirectional) randomized as the vertical strip with three dosages of each herbicide for a total of 12 wiping treatments. Data were collected at 35 and 90 DAT. The percent plant mortality was calculated using differences in pre- and posttreatment plant counts. ANOVA and log-logistic linear regression were used to analyze the data. The dose–response experiment showed that S. indicus var. pyramidalis mortality increased with herbicide concentration, and mortality was greater with the bidirectional wiping method compared with the unidirectional method. Treatments wiped bidirectionally with glyphosate at 70% v/v, hexazinone at 30% v/v, and hexazinone at 60% v/v resulted in S. indicus var. pyramidalis mortality ranging from 75% to 98% by 90 DAT across all locations. The ATV-mounted weed-wiper experiment showed that mowing before herbicide application with weed wipers decreased the efficacy of both herbicides. Overall, both experiments indicate that S. indicus var. pyramidalis should be wiped bidirectionally using either glyphosate (70% v/v) or hexazinone (at least 30% v/v) to obtain satisfactory control. Further work should be conducted to determine whether seasonality impacts the response of S. indicus var. pyramidalis to mowing and the application of these herbicides.

On the Occurrence of Two Species of <i>Hypochoeris</i> Linn. in Nilgiris, South India

A chromosomal count of Hypochoeris plants collected at Ootacamund indicated the identity of those plants as H. radicata L., thus pointing to its presence in the Nilgiris, in addition to H. glabra L. reported in earlier works. This is confirmed by a detailed morphological study of living plants and herbarium specimens. A gradation in morphological characters between the two species leading to a possibility of interspecific hybridization is suggested. This may also account for the earlier misidentification of H. radicata with H. glabra. A table showing distinctive characters of the two species is provided.

The Effect of Azotobacter on Growth and Yield of Water Spinach in Growing Media Containing Gold Mine Tailings

Gold mining provides economic benefits but results in the conversion of agricultural land and produces low-fertility tailings. Efforts to remediate tailings can be done through the utilization of the biofertilizer bacteria Azotobacter sp., which is capable of producing exopolysaccharides. This study aims to study the effects of biofertilizers containing a consortium of Azotobacter sp. strains S5, S6a, and S9 in enhancing the growth and yield of water spinach plants in growth media containing gold mine tailings. The experiment used a randomized complete block design with four treatments and six replications. The treatments consisted of; (A) control, (B) 1% biofertilizer, (C) 5% biofertilizer, and (D) 10% biofertilizer. The results of the experiment showed that the dosage of 5% biofertilizers increased plant height, leaf count, and fresh weight of plants, while the 10% biofertilizer increased the stem diameter. The findings of this experiment emphasize the importance of biofertilizers in enhancing the growth and yield of water spinach plants in media with low fertility.

Rice Counting and Localization in Unmanned Aerial Vehicle Imagery Using Enhanced Feature Fusion

In rice cultivation and breeding, obtaining accurate information on the quantity and spatial distribution of rice plants is crucial. However, traditional field sampling methods can only provide rough estimates of the plant count and fail to capture precise plant locations. To address these problems, this paper proposes P2PNet-EFF for the counting and localization of rice plants. Firstly, through the introduction of the enhanced feature fusion (EFF), the model improves its ability to integrate deep semantic information while preserving shallow spatial details. This allows the model to holistically analyze the morphology of plants rather than focusing solely on their central points, substantially reducing errors caused by leaf overlap. Secondly, by integrating efficient multi-scale attention (EMA) into the backbone, the model enhances its feature extraction capabilities and suppresses interference from similar backgrounds. Finally, to evaluate the effectiveness of the P2PNet-EFF method, we introduce the URCAL dataset for rice counting and localization, gathered using UAV. This dataset consists of 365 high-resolution images and 173,352 point annotations. Experimental results on the URCAL demonstrate that the proposed method achieves a 34.87% reduction in MAE and a 28.19% reduction in RMSE compared to the original P2PNet while increasing R2 by 3.03%. Furthermore, we conducted extensive experiments on three frequently used plant counting datasets. The results demonstrate the excellent performance of the proposed method.

Analyzing Winter Wheat (Triticum aestivum) Growth Pattern Using High Spatial Resolution Images: A Case Study at Lakehead University Agriculture Research Station, Thunder Bay, Canada

Remote sensing technology currently facilitates the monitoring of crop development, enabling detailed analysis and monitoring throughout the crop’s growing stages. This research analyzed the winter wheat growth dynamics of experimental plots at the Lakehead University Agricultural Research Station, Thunder Bay, Canada using high spatial and temporal resolution remote sensing images. The spectral signatures for five growing stages were prepared. NIR reflectance increased during the growing stages and decreased at the senescence, indicating healthy vegetation. The space–time cube provided valuable insight into how canopy height changed over time. The effect of nitrogen treatments on wheat did not directly influence the plant count (spring/autumn), and height and volume at maturity. However, the green and dry weights were different at several treatments. Winter wheat yield was predicted using the XGBoost algorithm, and moderate results were obtained. The study explored different techniques for analyzing winter wheat growth dynamics and identified their usefulness in smart agriculture.

Implementation of a Wireless Sensor Network for Environmental Measurements

Nowadays, the need to monitor different physical variables is constantly increasing and can be used in different applications, from humidity monitoring to disease detection in living beings, using a local or wireless sensor network (WSN). The Internet of Things has become a valuable approach to climate monitoring, daily parcel monitoring, early disease detection, crop plant counting, and risk assessment. Herein, an autonomous energy wireless sensor network for monitoring environmental variables is proposed. The network’s tree topology configuration, which involves master and slave modules, is managed by microcontrollers embedded with sensors, constituting a key part of the WSN architecture. The system’s slave modules are equipped with sensors for temperature, humidity, gas, and light detection, along with a photovoltaic cell to energize the system, and a WiFi module for data transmission. The receiver incorporates a user interface and the necessary computing components for efficient data handling. In an open-field configuration, the transceiver range of the proposed system reaches up to 750 m per module. The advantages of this approach are its scalability, energy efficiency, and the system’s ability to provide real-time environmental monitoring over a large area, which is particularly beneficial for applications in precision agriculture and environmental management.

Nitrogen application and cultivar effects on industrial hemp yield dynamics

AbstractRenewed interest in hemp (Cannabis sativa L., <0.3% delta‐9‐tetrahydrocannabinol) production has motivated investigations into crop management and fertilizer application. Due to fertilizer regulations, soil characteristics, and unique climatic conditions in Florida, researching nutrient management for region‐specific hemp cultivation is essential. A study conducted in central Florida during 2021 and 2022 used six nitrogen (N) rates (0, 56, 112, 168, 224, and 280 kg ha−1) and five cultivars (NWG‐2730, X‐59, Bialobrzeskie, IH Williams, and Yuma) as a first effort to develop producer‐oriented fertilizer guidelines. The field experiments employed a randomized complete block design across N rates and varieties, with N applied in three split applications. The results indicated an interaction between cultivars and N rates, as expected due to their diverse genetic and geographic origins. Challenges in genetic and germination quality resulted in grain and biomass yield of two cultivars, X‐59 and Bialobrzeskie, that did not respond to the N application. Conversely, NWG‐2730, IH Williams, and Yuma cultivars demonstrated significantly higher yields within the 168–280 kg ha−1 N range. Moreover, cultivars demonstrated differing emergence counts (field count), and findings indicated a correlation between emergence and plant count at harvest (r = 0.68 and p = 0). The study also found a difference in germination count versus field count (after adjusting the seed rate depending on germination percentage). Enhanced plant count at harvest substantially elevated grain and biomass yield. The N application positively influenced plant height across all cultivars except Bialobrzeskie. Optimal plant height was achieved within the 168–280 kg N ha−1 range. Therefore, optimizing N rates and selecting cultivars (with higher emergence rates in the field) could help achieve higher biomass and grain yields.

Quantifying consistency of crop establishment using a lightweight U-Net deep learning architecture and image processing techniques

Consistency of crop establishment is a measure of uniformity of crop attributes, such as plant stand count, crop emergence rate, and plant spacing across the field. Quantifying consistency during the early crop growth stage is important for establishment decisions to use targeted nutrients and to facilitate timely replanting in inconsistent crop regions. Crop consistency can be analysed using two key parameters: plant stand count and spacing statistics since they provide insight into plant density and its emergence percentage. However, manual assessment of them is time-consuming, prone to errors, and labour-intensive in large fields. An alternative method is proposed to automate estimating these parameters using field imagery under uncontrolled settings. We use the YOLOv5-based object detection model for plant counting, which attains a mean average precision of 0.956 to detect Canola plants. A Lightweight U-Net model is proposed to segment rows, followed by Guo–Hall thinning and Probabilistic Hough Transform to determine inter-row and inter-plant spacing. Our proposed row segmentation model achieves a mean Intersection over Union (mIoU) of 0.8444 with class-wise IoU of 0.9925 and 0.6963 for background and crop using fewer parameters. The new architecture uses only 14M parameters and achieves performance comparable to the state-of-the-art U-Net (32.5M) and SegNet (29M).

Aerial imagery-based tobacco plant counting framework for efficient crop emergence estimation

Crop emergence estimation at early crop growth stages is becoming increasingly important for the long-term sustainability of natural resources. It helps farmers and agricultural stakeholders in the efficient allocation of resources like water, pesticides, and fertilizers. It can be used to estimate the yield and seed quality, identify the region of potential yield losses, and make future agriculture plans. These future agriculture plans can play a crucial role in ensuring maximum crop population and yield while utilizing the same limited land and natural resources. Most of the existing plant counting frameworks require offline processing of images with computationally expensive algorithms including the structure for motion and multiview stereo to develop an orthomosaic. This study proposed a tobacco plant counting framework that directly estimates counts from aerial images and has the potential for real-time applicability. It consists of three core modules: overlap detection, plant detection, and plant counting. The overlap detection module replaces the need for computationally expensive orthomosaic formation to avoid counting repetition by overlap masking based on only visual cues. Three different methods are evaluated as core modules for finding an optimal solution for plant counting based on time complexity and accuracy. In the first method after overlap detection, semantic segmentation with U-NET is employed as a plant detection module. For plant counting, we count the connected pixels classified as plants to estimate the crop count. In the second method after overlap detection, object detection using YOLOv7 is utilized as a plant detection module followed by simply counting each detected plant. In the third method, we utilize YOLOv7 for object detection, similar to the second method. However, we introduce the SORT (Simple Online and Realtime Tracking) algorithm for object tracking. This object tracking replaces the overlap detection module making it a real-time applicable method. For plant counting, we assess the number of tracked plants. The proposed algorithm is evaluated on two distinct tobacco fields. The high-resolution aerial data is collected from tobacco fields near Peshawar, Pakistan, and is human-labelled. The first and second methods show average F1 scores of 0.947 and 0.9667, respectively, whereas the third method has the potential for real-time applicability with an average F1 score of 0.967.

Corn Plant Counting Using Deep Learning and UAV Images

The adoption of new technologies, such as unmanned aerial vehicles (UAVs), image processing, and machine learning, is disrupting traditional concepts in agriculture, with a new range of possibilities opening in its fields of research. Plant density is one of the most important corn (Zea mays L.) yield factors, yet its precise measurement after the emergence of plants is impractical in large-scale production fields due to the amount of labor required. This letter aims to develop techniques that enable corn plant counting and the automation of this process through deep learning and computational vision, using images of several corn crops obtained using a low-cost unmanned aerial vehicle (UAV) platform assembled with an RGB sensor.

A point-in-time inventory of chikanda orchids within a wild harvesting wetland area in Mwinilunga, Zambia: implications for conservation

Despite continued wild harvesting, there is limited data that provides estimates on the availability of African edible orchids as a resource, and that identifies the ecological drivers of their dynamics. This study was designed to investigate the relationship between distance from surface water and the diversity, frequency, and population density of chikanda orchids in a harvesting wetland in Mwinilunga, Zambia. Vegetation sampling was conducted using an interrupted belt transect design. Eight 50 m transects were systematically placed and positioned perpendicular (90o) to a stream within three separate orchid communities. Individual plant count was recorded within the 0–10 m, 10–20 m, 20–30 m, 30–40 m, and 40–50 m distance zones. Results showed a statistically significant decrease in the Shannon’s Diversity Index between the distance zones in all three communities (p < 0.05). Mean ranks for orchid taxa frequency and density across the distance zones were significantly different for communities 1 and 2 but not 3 and communities 1 and 3 but not 2, respectively. However, there was a medium to large effect size for distance on frequency and a small to large effect size for distance on density in all three communities. These results suggest that distance from surface water influences the frequency and density of orchid taxa occurring in wetland ecosystems. This implies that for conservation purposes, it may be important to focus on areas close to water when establishing orchid sites for the species investigated in this study.

Method for maize plants counting and crop evaluation based on multispectral images analysis

The processing of multispectral images acquired with embedded cameras in unmanned aerial vehicles (drones) has brought new opportunities for precision agriculture. In this study a method for evaluating the number of corn plants (Zea mays L) in a crop area is presented. Plant density is one of the most important yield factors, yet its precise measurement after the emergence of plants is impractical in large and medium-scale production, since significant amount of labor is required. For validation, a dataset of spectral images was gathered from flights over an agricultural area, and digital image processing techniques were applied, taking into account the concept of intelligent processing. Therefore, pattern recognition and models to aid decision-making through machine learning were also used. After image acquisition, the processing of orthomosaics in the spectral channels, i.e., red (R), green (G), and blue (B), was performed, making it possible to register and organize all the images. Likewise, techniques for geometric transformation, brightness, and contrast adjustments were evaluated globally, whereas local adjustments were evaluated based on the use of adaptive equalization techniques, which were explored based in the choice of the HSV color space. For the post-processing step, segmentation based on the best observed color threshold technique, in conjunction with Gaussian filtering and morphological operations, were considered. To enable pattern recognition, techniques that use distance maps were evaluated, considering the use of Euclidean distance. Thus, the locations of canopy patterns in maize plants were studied using a template matching algorithm and Chamfer pattern mask. For feature extraction, chain code and circular pattern map techniques were considered. The analyses made it possible to establish vectors of features based on patterns related to the number of maize plants occurrences. Finally, three calibration steps were considered, one related to the plant height versus the canopy opening radius, other related to the number of maize plants for each position in the crop area versus the radii identified by the developed model, and the third related to the cross-correlation between the plant counting by human vision and the new method. In addition, the classification step was established using a set of classifiers based on support vector machine (SVM). Results have shown an accurate and timely counting methodology for maize plants, which can guide cultivation to ensure high yield. The results showed that as a new method it can effectively count the number of maize plants with an average accuracy rate equal to 88.47%. Besides, both selected SVM classifiers have presented accuracy higher than 84% and precision higher than 83%. Furthermore, the cross-correlation between the plant counting by human vision and the new method has presented a linear correlation coefficient equal to 0.98. Thus, the developed method proved to be adequate for counting the maize plants in the post-emergence stage.

Automated Counting of Tobacco Plants Using Multispectral UAV Data

Plant counting is an important part in precision agriculture (PA). The Unmanned Aerial Vehicle (UAV) becomes popular in agriculture because it can capture data with higher spatiotemporal resolution. When it is equipped with multispectral sensors, more meaningful multispectral data is obtained for plants’ analysis. After tobacco seedlings are raised, they are transplanted into the field. The counting of tobacco plant stands in the field is important for monitoring the transplant survival rate, growth situation, and yield estimation. In this work, we adopt the object detection (OD) method of deep learning to automatically count the plants with multispectral images. For utilizing the advanced YOLOv8 network, we modified the architecture of the network to adapt to the different band combinations and conducted extensive data pre-processing work. The Red + Green + NIR combination obtains the best detection results, which reveal that using a specific band or band combinations can obtain better results than using the traditional RGB images. For making our method more practical, we designed an algorithm that can handling the image of a whole plot, which is required to be watched. The counting accuracy is as high as 99.53%. The UAV, multispectral data combined with the powerful deep learning methods show promising prospective in PA.