The automatic identification of grassland plants based on UAV remote sensing imagery is crucial for the management and conservation of grassland ecosystems. Grassland plants, characterized by small individual sizes, complex backgrounds, and dense distribution, pose a challenge for UAV image identification. An improved YOLOv8 algorithm is proposed in this study to enhance the detection effect of grassland plants from different angles and optimize the detection of small targets in complex backgrounds. Firstly, a rotation-invariant mechanism is introduced into the backbone to improve the model’s adaptation to grassland plant targets from various angles, thereby improving detection accuracy and robustness. Secondly, a self-attention mechanism is applied in the backbone to enhance the model’s understanding capability of correlations between different plants. Additionally, a multi-level feature extraction structure based on dilated convolution is employed in the neck to further enhance the model’s feature representation capability. Finally, the SlideLoss loss function is introduced to solve the problems of insufficient sample size and foreground-background detection impact. Experimental results demonstrate that the improved algorithm achieves remarkable results in the automatic identification of grassland plants in Inner Mongolia, validating the application prospects of UAV remote sensing technology in grassland plant species monitoring.
Graphite is one of the important industrial mineral raw materials, but the high content of heavy metals in tailings may cause soil pollution and other regional ecological environmental problems. Luobei has already become the largest production base of graphite. To find out the ecological situation in the region, further ecological risk analysis has been carried out. Luobei graphite mine which is located in Yabdanhe basin has been selected as the study area, SVM classifiers method with the support of GF-1 Satellite remote sensing data has been used, which is the first high-resolution earth observation satellite in China. The surrounding ecological environment was monitored and its potential impact on the ecological environment was analyzed by GIS platform. The results showed that the Luobei graphite mine located Yadanhe basin covers an area of 499.65 km2, the main types of forest ecosystems ( 44.05% of the total basin area ), followed by agricultural area( 35.14% ), grass area( 15.52% ), residential area ( 4.34% ), mining area ( 0.64% ) and water area( 0.30% ). By confirming the classification results, the total accuracy is 91.61%, the Kappa coefficient is 0.8991. Overall, GF-1 Satellite data can obtain regional ecosystems quickly, and provide a better data support for regional ecological resource protection zone. For Luobei graphite mines area, farmland and residential areas within its watershed are most vulnerable to mining, the higher proportion of farmland in duck river basin. The regulatory tailings need to be strengthened in the process of graphite mining processing.
Radiosity method is based on the computer simulation of 3D real structures of vegetations, such as leaves, branches and
stems, which are composed by many facets. Using this method we can simulate the canopy reflectance and its
bidirectional distribution of the vegetation canopy in visible and NIR regions. But with vegetations are more complex,
more facets to compose them, so large memory and lots of time to calculate view factors are required, which are the
choke points of using Radiosity method to calculate canopy BRF of lager scale vegetation scenes. We derived a new
method to solve the problem, and the main idea is to abstract vegetation crown shapes and to simplify their structures,
which can lessen the number of facets. The facets are given optical properties according to the reflectance, transmission
and absorption of the real structure canopy. Based on the above work, we can simulate the canopy BRF of the mix scenes
with different species vegetation in the large scale. In this study, taking broadleaf trees as an example, based on their
structure characteristics, we abstracted their crowns as ellipsoid shells, and simulated the canopy BRF in visible and NIR
regions of the large scale scene with different crown shape and different height ellipsoids. Form this study, we can
conclude: LAI, LAD the probability gap, the sunlit and shaded surfaces are more important parameter to simulate the
simplified vegetation canopy BRF. And the Radiosity method can apply us canopy BRF data in any conditions for our research.
In this paper, a portable diagnostic instrument was designed and tested, which can measure the normalized difference vegetation index (NDVI) and structure insensitive pigment index (SIPI) of crop canopy in field. The instrument have a valid survey area of 1 m*1 m when the height between instrument and the ground was fixed to 1.3 meter The crop growth condition can be assessed based on their NDVI and SIPI values, so it will be very important for crop management to get these values. The instrument uses sunlight as its light source. There are six special different photoelectrical detectors within red, blue and near infrared bands, which are used for detecting incidence sunlight and reflex light from the canopy of crop. This optical instrument includes photoelectric detector module, signal process and A/D convert module, the data storing and transmission module and human-machine interface module. The detector is the core of the instrument which measures the spectrums at special bands. The microprocessor calculates the NDVI and SIPI value based on the A/D value. And the value can be displayed on the instrument's LCD, stored in the flash memory of instrument and can also be uploaded to PC through the PC's RS232 serial interface. The prototype was tested in the crop field at different view directions. This paper also provided the method of calibration, the results showed that the average measurement error to SIPI value of instrument was 5.25% and the average measurement error to NDVI value in vegetation-covered region is 6.40%. It reveals the on-site and non-sampling mode of crop growth monitoring by fixed on the agricultural machine traveling in the field.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.