• Agricultural and Biosystems Engineering
• /
• v.6 no.1
• /
• pp.1-7
• /
• 2005

Assessing health condition of crop in the field is one of core operation in precision fanning. A sensing system was proposed to remotely detect the crop health condition in terms of SP AD readings directly related to chlorophyll contents of crop using a multispectral camera equipped on ground-based platform. Since the image taken by a camera was sensitive to changes in ambient light intensity, it was needed to convert gray scale image data into reflectance, an index to indicate the reflection characteristics of target crop. A reference reflectance panel consisting of four pieces of sub-panels with different reflectance was developed for a dynamic calibration, by which a calibration equation was updated for every crop image captured by the camera. The system performance was evaluated in a field by investigating the relationship between com canopy reflectance and SP AD values. The validation tests revealed that the com canopy reflectance induced from Green band in the multispectral camera had the most significant correlation with SPAD values $(r^2=0.75)$ and NIR band could be used to filter out unwanted non-crop features such as soil background and empty space in a crop canopy. This research confirmed that it was technically feasible to develop a ground-based remote sensing system for assessing crop health condition.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 1998.04a
• /
• pp.133-134
• /
• 1998

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.10a
• /
• pp.81-81
• /
• 2022

Recently, many studies on big data based smart farming have been conducted. Research to quantify morphological characteristics using image data from various crops in smart farming is underway. Rice is one of the most important food crops in the world. Much research has been done to predict and model rice crop yield production. The number of productive tillers per plant is one of the important agronomic traits associated with the grain yield of rice crop. However, modeling the basic growth characteristics of rice requires accurate data measurements. The existing method of measurement by humans is not only labor intensive but also prone to human error. Therefore, conversion to digital data is necessary to obtain accurate and phenotyping quickly. In this study, we present an image-based method to predict leaf number and evaluate tiller number of individual rice crop using YOLOv5 deep learning network. We performed using various network of the YOLOv5 model and compared them to determine higher prediction accuracy. We ako performed data augmentation, a method we use to complement small datasets. Based on the number of leaves and tiller actually measured in rice crop, the number of leaves predicted by the model from the image data and the existing regression equation were used to evaluate the number of tillers using the image data.

• Korean Journal of Remote Sensing
• /
• v.29 no.6
• /
• pp.611-619
• /
• 2013

Although the need of crop production has increased in Mongolia, crop cultivation is very limited because of the harsh climatic and topographic conditions. Crop lands are sparsely distributed with relatively small sizes and, therefore, it is difficult to survey the exact area of crop lands. The study aimed to find an easy and effective way of accurate classification to map crop lands in Mongolia using satellite images. To classify the crop lands over the study area in northern Mongolia, four classifications were carried out by using 1) Thematic Mapper (TM) image August 23, 2) TM image of July 6, 3) combined 12 bands of TM images of July and August, and 4) both TM images of July and August by layered classification. Wheat and potato are the major crop types and they show relatively high variation in crop conditions between July and August. On the other hands, other land cover types (forest, riparian vegetation, grassland, water and bare soil) do not show such difference between July and August. The results of four classifications clearly show that the use of multi-temporal images is essential to accurately classify the crop lands. The layered classification method, in which each class is separated by a subset of TM images, shows the highest classification accuracy (93.7%) of the crop lands. The classification accuracies are lower when we use only a single TM image of either July or August. Because of the different planting practice of potato and the growth condition of wheat, the spectral characteristics of potato and wheat cannot be fully separated from other cover types with TM image of either July or August. Further refinements on the spatial characteristics of existing crop lands may enhance the crop mapping method in Mongolia.

• Journal of Digital Contents Society
• /
• v.19 no.6
• /
• pp.1197-1205
• /
• 2018

In this paper, we perform Single Image Super Resolution(SISR) for acquired images of EOTS or IRST from naval combat system. In order to conduct super resolution, we use Generative Adversarial Networks(GANs), which consists of a generative model to create a super-resolution image from the given low-resolution image and a discriminative model to determine whether the generated super-resolution image is qualified as a high-resolution image by adjusting various learning parameters. The learning parameters consist of a crop size of input image, the depth of sub-pixel layer, and the types of training images. Regarding evaluation method, we apply not only general image quality metrics, but feature descriptor methods. As a result, a larger crop size, a deeper sub-pixel layer, and high-resolution training images yield good performance.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.10a
• /
• pp.80-80
• /
• 2022

Estimation of heading date of paddy rice is laborious and time consuming. Therefore, automatic estimation of heading date of paddy rice is highly essential. In this experiment, deep learning classification models were used to classify two difference categories of rice (vegetative and reproductive stage) based on the panicle initiation of paddy field. Specifically, the dataset includes 444 slanted view images belonging to two categories and was then expanded to include 1,497 images via IMGAUG data augmentation technique. We adopt two transfer learning strategies: (First, used transferring model weights already trained on ImageNet to six classification network models: VGGNet, ResNet, DenseNet, InceptionV3, Xception and MobileNet, Second, fine-tuned some layers of the network according to our dataset). After training the CNN model, we used several evaluation metrics commonly used for classification tasks, including Accuracy, Precision, Recall, and F1-score. In addition, GradCAM was used to generate visual explanations for each image patch. Experimental results showed that the InceptionV3 is the best performing model in terms of the accuracy, average recall, precision, and F1-score. The fine-tuned InceptionV3 model achieved an overall classification accuracy of 0.95 with a high F1-score of 0.95. Our CNN model also represented the change of rice heading date under different date of transplanting. This study demonstrated that image based deep learning model can reliably be used as an automatic monitoring system to detect the heading date of rice crops using CCTV camera.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.22 no.1
• /
• pp.13-19
• /
• 2020

This study compared RGB color images with canopy light interception (LI) and leaf area index (LAI) measurements for low cost and low labor. LAI and LI were measured from vertical gap fraction derived from top of digital image in soybean canopy cover (cv Daewonkong, Deapongkong and Pungsannamulkong). RGB color images, LAI, and LI were collected from V4.5 stage to R5stage. Image segmentation was based on excess green minus excess red index (ExG-ExR). There was a linear relationship between LAI measured with LI (r²=0.84). There was alinear relation ship between LI measured with canopy cover on image (CCI) (r²=0.94). There was a significant positive relationship(r²=0.74) between LAI and CCI at all grow ingseason. Therefore, it is expected that in the future, the RGB color image could be able to easily measure the LAI and the LI at low cost and low labor.

• Journal of Biosystems Engineering
• /
• v.24 no.1
• /
• pp.25-30
• /
• 1999

The effectiveness of many greenhouse environment control methodologies depends on the growth information of crops. Acquisition of the growth information of crops requires a non-invasive and continuous monitoring method. Crop growth monitoring system using digital imaging technique was developed to conduct non-destructive and intact plant growth analyses. The monitoring system automatically measures crop growth information sends an appropriate control signal to the nutrient solution supplying system. To develop the monitoring system, a linear model that explains the relationship between the fresh weight and the top projected leaf area of a lettuce plant was developed from an experiment. The monitoring system was evaluated buy successive lettuce growing experiments. Results of the experiments showed that the developed system could estimate the fresh weight of lettuce from a lettuce image by using the linear model and generate an EC control signal according to the lettuce growth stage.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.10a
• /
• pp.108-108
• /
• 2022

In this study, multispectral images of wheat according to soil water state were collected, compared, and analyzed to measure the physiological response of crops to environmental stress at the individual level. CMS-V multi-spectral camera(Silios Technologies) was used for image acquisition. The camera lens consists of eight spectral bands between 550nm and 830nm. Light Reflective information collected in each band sensor and stored in digital values, and it is converted into a reflectance for calculating the vegetation index and used. According to the camera manual, the NDVI(Normalized Difference vegetation index) value was calculated using 628 nm and 752 nm bands. Image measurement was conducted under natural light conditions, and reflectance standards(Labsphere) were captured with plants for reflectance calculation. The wheat variety used Gosomil, and the wheat grown in the field was transplanted into a pot after heading date and measured. Three treatments were performed so that the soil volumetric water content of the pot was 13~17%, 20~23%, and 25%, and the growth response of wheat according to each treatment was compared using the NDVI value. In the first measurement after port transplantation, the difference in NDVI value according to treatment was not significant, but in the subsequent measurement, the NDVI value of the treatment with a water content of 13 to 17% was lowest and was the highest at 20 to 23%. The NDVI values decreased compared to the first measurement in all treatment, and the decrease was the largest at 13-17% water content and the smallest at 20-23%. Although the difference in NDVI values could be confirmed, it would be difficult to directly relate it to the water stress of plants, and further research on the response of crops to environmental stress and the analysis of multi-spectral image will be needed.

• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.149-158
• /
• 2024

Traditional manual identification of crop leaf diseases is challenging. Owing to the limitations in manpower and resources, it is challenging to explore crop diseases on a large scale. The emergence of artificial intelligence technologies, particularly the extensive application of deep learning technologies, is expected to overcome these challenges and greatly improve the accuracy and efficiency of crop disease identification. Crop leaf disease identification models have been designed and trained using large-scale training data, enabling them to predict different categories of diseases from unlabeled crop leaves. However, these models, which possess strong feature representation capabilities, require substantial training data, and there is often a shortage of such datasets in practical farming scenarios. To address this issue and improve the feature learning abilities of models, this study proposes a deep transfer learning adaptation strategy. The novel proposed method aims to transfer the weights and parameters from pre-trained models in similar large-scale training datasets, such as ImageNet. ImageNet pre-trained weights are adopted and fine-tuned with the features of crop leaf diseases to improve prediction ability. In this study, we collected 16,060 crop leaf disease images, spanning 12 categories, for training. The experimental results demonstrate that an impressive accuracy of 98% is achieved using the proposed method on the transferred ResNet-50 model, thereby confirming the effectiveness of our transfer learning approach.