Efficient disaster surveys can safeguard the compensation rights of affected farmers and serve as a critical component of market stabilization strategies. Due to climate change, Taiwan is experiencing more frequent climate disasters. The agricultural industry faces severe labor shortages making post-disaster recovery increasingly challenging. This issue may impact food supply security, disrupt prices, and threaten national security. To address this, the study applies an advanced and efficient semantic segmentation network model to the Unmanned Aerial Vehicle (UAV) captured imagery for rice lodging disaster assessment. By incorporating a rule-based multi-task learning framework, prior knowledge from physical rules constrains the classifier's learning. Preliminary results indicate that the modified model achieved a 10% above improvement in the recall rate for lodged rice compared to the original model using 2017 data, and around a 5% improvement on the transferred 2019 data. Suggesting that this study can predict rice lodging with a more interpretable model architecture and achieve better classification results.
In traditional unsupervised classification method, the number of clusters usually needs to be assigned subjectively by
analysts, but in fact, in most situations, the prior knowledge of the research subject is difficult to acquire, so the suitable
and best cluster numbers are very difficult to define. Therefore, in this research, an effective heuristic unsupervised
classification method-Genetic Algorithm (GA) is introduced and tested here, because it can be through the
mathematical model and calculating procedure of optimization to determine the best cluster numbers and centers
automatically. Furthermore, two well-known models--Davies-Bouldin's and the K-Means algorithm, which adopted by
most research for the applications in pattern classification, are integrated with GA as the fitness functions. In a word, in
this research, a heuristic method-Genetic Algorithm (GA), is adopted and integrated with two different indices as the
fitness functions to automatically interpret the clusters of satellite images for unsupervised classification. The
classification results were compared to conventional ISODATA results, and to ground truth information derived from a
topographic map for the estimation of classification accuracy. All image-processing program is developed in MATLAB,
and the GA unsupervised classifier is tested on several image examples.
Traditionally, an unsupervised classification divides all pixels within an image into a corresponding class pixel by pixel.
The number of classes must be selected, but seldom is ascertainable with little information in advance. Moreover,
spectral properties of specific informational classes change seasonally for satellite imagery. The relationships between
informational classes and spectral classes are not always constant, and relationships defined for one image cannot be
extended to others. Thus, the analyst has very limited or no control over the menu of classes and their specific identities.
In this study, a Genetic Algorithm is adopted to interpret the cluster centers of an image and to reveal a suitable number
of classes to overcome the disadvantage of unsupervised classification. A Genetic Algorithm is capable of dealing with
a set of numerous data such as satellite imagery pixels. An optimization consequence of the image classification is
introduced and carried out. Through an image process program developed in Mathlab, the GA unsupervised classifier
was processed on several test images for validity and on SPOT satellite imagery. The classified SPOT image was
compared with finer aerial photographs as a ground truth for the estimation of classification accuracy.
Eutrophication is one of the most common problems of water resources in developed and developing countries. Traditional measurement of water quality requires on-site sampling and laboratory work, which is expensive and time consuming. Due to these imitations, the sample size is often too small to have a high reliability of the corresponding results especially for a large water body. Remote sensing provides a new technique to monitor water quality over a wide area with a two-dimensional data distribution instead of sample points. In this research, French satellite SPOT was chosen as remotely sensed data source and provided images to derive chlorophyll concentration, Secchi depth, and phosphorous concentration for a water body. By comparing a set of on-site samples and the corresponding brightness values on a SPOT image taken on the same date, a regression model converting satellite data to water quality variables was defined. A systematic image process was developed to transfer SPOT data to water quality variables. This system provides not only an instantaneous and repetitive eurtrophic status assessment but also a visualizing water quality variation. An image process and GIS software IMAGINE was adopted to carry out the process in a case study of the Te-Chi Reservoir in Taiwan. The final product is a set of thematic maps of eutrophic status (represented by Carlson's TSI) of the reservoir.
Conference Committee Involvement (1)
Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques, and Applications VIII
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.