Language：Japanese   Publisher：The Japanese Agricultural Systems Society  

<p>Plant height, or the natural height from the ground surface to the top of a plant, is an essential metric for estimating the growth status of cultivated vegetables. The variation of plant height over time indicates the intensity of plant vigor, determining the amount of irrigation and fertilizer application during cultivation. Recently, various 3D-shape measuring instruments have been used for automatic plant height measurement, among which the RGB-D camera is reported to be relatively inexpensive and easy to acquire depth data. The previous study proposed an RGB-D camera-based plant height estimation method that assesses the plant height by averaging the 5% and 10% uppermost pixels. However, some unexpected inclusion into the pixels with the leaves closing to the top of stems lowers the estimated height, resulting in low accuracy (84%~100% and 45%~100%, respectively). This paper presents a novel algorithm to search the tops of stems to estimate the plant height. An RGB-D camera was placed directly above the target plants to continuously take pictures from the planting stage when the density of leaves was relatively small. The algorithm marked the highest pixel as the first top of plant stems and excluded a specific range around it before continuing the search, aiming to eliminate the influence from the lower leaves. This process was repeated until the number of found tops agreed with the number of stems; the plant height was then calculated by averaging the height of stem tops. As an evaluation experiment, the estimated height was compared with the measured value, showing a high accuracy of 94%~100% in our method.</p>

DOI： 10.14962/jass.38.1_15

Publisher：The Japanese Agricultural Systems Society  

<p>In recent years, the increased occurrence of sudden heavy rains in Japan has increased the occurrences of surface runoff from sloping farmlands. These runoff events negatively affect the ecosystems owing to the leaching of fertilizer components and/or heavy metals into local watersheds. Parshall flumes have been conventionally used for measuring the extent of surface runoff combined with water-level gauges installed at the bottom edge of sloping farmlands. However, data obtained using this equipment are unreliable because of the direct contact of sensors with various objects, such as pile-ups of crop residues and/or soil particles. To avoid this problem, we propose a novel method herein. Firstly, we devised a time-lapse camera which shot the inside-view images of the Parshall flume and used automatically captured time-lapse images of water flow running through the Parshall flume to estimate the precise runoff extent by detecting the instantaneous water depth using image processing techniques. Second, we devised an automated video-capturing system and positioned it 1.8 m above the ground in an outdoor artificial sloping field in Kagoshima, Japan. Vertically captured video-images of the surface runoff were obtained during a severe runoff event. We analyzed the captured video-images using particle tracking velocimetry to determine the velocity of the moving surface water. The widths of surface water flow were also measured on the video-images. We estimated the runoff amounts using the velocity and the flow width, in addition to the given slope gradient and a literature value of the equivalent roughness coefficient for the soil. The results of two techniques are comparable, especially at severe rainfall and runoff occasions, in which one was measured using the Parshall flume with time-lapse images and the other was measured using the vertically captured video-images. The proposed method may provide estimates of the amount of the surface runoff during intense rainfall events solely using video-images.</p>

DOI： 10.14962/jass.35.4_69

Publisher：The Japanese Agricultural Systems Society  

<p>We demonstrated a close-range sensing technique to seasonally track trends in canopy leaf greenness in rice paddies. A weatherproof narrow-band triplet digital camera system was positioned 3 m above a paddy rice field during the 2017 rice season in Kagoshima. The system automatically logged reflectance images at 10-min intervals from 8:00 to 16:00 in visible green （560 ± 10 nm）, red （650 ± 10 nm）, and near-infrared （855 ± 10 nm） bands. Two rice cultivars were grown at two fertilizer concentrations and seedling numbers in three replicates in concrete-framed plots. The camera was mounted on a platform at the top of an iron pole installed on a handcart built with scaffolding pipes. The east-looking camera captured oblique images of four plots （2 × 2 plots） with the camera's axis tilted 65^o from vertical with the handcart positioned 3.2 m away from the west side of the plots. To capture images of a different group of plots each day, the handcart was manually moved to a randomly chosen set position among the five predetermined positions north and south along the field on the concrete-paved open side area every morning. Radiometric corrections for images utilized solar irradiance sensors and prior calibrations to calculate instantaneous reflectance in each band. The leaf greenness value （SPAD） and the leaf area index （LAI） were weekly collected for rice canopies, and daily values were obtained by interpolating the weekly data. From reflectance images, SPAD and LAI data were analyzed for 35-day period including the panicle formation stage. The canopy greenness index, defined as the product of （SPAD – 30） and LAI （which potentially represents canopy nitrogen condition）, highly correlated （r = 0.62, n = 4276） with the canopy surface averages of the bi-band ratio （NIR/Green）, where NIR and Green indicate the reflectance values in the near-infrared and green band, respectively. Further investigations including daily-averaged or hourly-sampled reflectance and/or bi-band ratios revealed that the correlation coefficient for the daily-averaged reflectance was 0.71(n = 128) and that for the hourly-sampled reflectance was 0.73 (n = 528) for the 15:00–16:00 period and 0.62 (n = 690) for the 11:00–14:00 period. We found that NIR/Green values obtained during the late afternoon time period best tracked the canopy greenness and differences in the cultivar and fertilization levels.</p>

DOI： 10.14962/jass.34.3_67

Publisher：The Association of Japanese Geographers  

The 1km mesh-size rice leaf wetness time forecast map was made by using relative humidity. This is effective for the predicton of rice-blast desease.

DOI： 10.14866/ajg.2018s.0_000066

Publisher：The Japanese Agricultural Systems Society  

<p>In recent years, the increased incidence of sudden heavy rains in Japan has enhanced the occurrences of surface runoff from sloping farmlands. These runoff events negatively affect the ecosystems due to leaching of fertilizer components and/or heavy metals into local watersheds. Water-level meters installed at the bottom edge of sloping farmlands have been conventionally used for measuring the extent of surface runoff. However, data provided by this equipment are unreliable because of the direct contact of sensors with various objects, such as pile-ups of crop residues and/or soil particles. To avoid this disadvantage, we herein propose a novel method that uses automatically captured time-lapse images of surface runoff. We devised an automated image-capturing system and have been using it in an outdoor artificial sloping field in Ibaraki Prefecture, Japan for over three months. Several sets of captured images, recorded actual rainfall levels, and runoff amounts were obtained during five typical runoff events caused by natural rainfalls. The captured images were analyzed at 10-minute intervals to determine the total volumes of water pools that were eventually formed with accumulated runoff water at the bottom of the sloping experimental field during each event. The estimates of accumulated water volume reasonably correlated with the instantaneous rainfall levels (r = 0.79***). Calculations were done for eight intense rainfall peaks during the observation period, assuming that the actual runoff amount could be estimated as the sum of the water amount measured with the water-level meter and the amount of accumulated water increased with time. We found that ratios of the amounts of surface runoff to rainfall water in each of the 10-minute intervals were within the range 12-59%. Our results are consistent with those of a previously published report of an experiment conducted at the same site but with a different method. The proposed method may provide more precise estimates of the amount of surface runoff during intense rainfall events.</p>

DOI： 10.14962/jass.33.1_11

Language：Japanese   Publishing type：Internal/External technical report, pre-print, etc.  

Language：Japanese   Publishing type：Internal/External technical report, pre-print, etc.  

Event date： 2019.3

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：静岡県立大学  

Event date： 2020.2

Language：Japanese  

Venue：東京都江戸川区総合文化センター  

Event date： 2019.10

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：喜界町役場  

Event date： 2024.7

Language：Japanese   Presentation type：Public lecture, seminar, tutorial, course, or other speech  

Venue：鹿児島市名山町  

Event date： 2023.12

Language：Japanese   Presentation type：Public lecture, seminar, tutorial, course, or other speech  

Venue：かごしま国際交流センター