Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Journal of Applied Geospatial Information Jurnal Penginderaan Jauh Indonesia
Muhammad Aldila Syariz
Institut Teknologi Sepuluh Nopember
Aerospace Engineering Agriculture, Biological Sciences & Forestry Civil Engineering, Building, Construction & Architecture Earth & Planetary Sciences Engineering Environmental Science Physics

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Estimasi Konsentrasi Klorofil-a menggunakan Refined Neural Network (Studi Kasus: Perairan Danau Kasumigaura) Aldila Syariz, Muhammad; Denaro, Lino Garda; Nabilah, Salwa; Heriza, Dewinta; Jaelani, Lalu Muhamad; Lin, Chao-Hung
Jurnal Penginderaan Jauh Indonesia Vol 1 No 1 (2019)
Publisher : Masyarakat Ahli Penginderaan Jauh Indonesia (MAPIN) /Indonesian Society of Remote Sensing (ISRS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (302.949 KB)

Abstract

Estimation of Chlorophyll-a Concentration using Refined Neural Network (Case Study: Lake Kasumigaura) Chlorophyll-a has been became one of clinical in-water constituents to represent water quality. Many researchers have used neural network method to estimate chlorophyll-a concentration in the water body. However, a few number of water samples limits the use of neural network, meaning that those number is insufficient to train the neural network model and makes the result is not reliable. One of famous interpolation method, that is Inverse Distance Weighting (IDW), is utilized in this study to enrich water samples dataset over non-station points. The data from those non-station points would further be used to train the neural network model. After the training, the neural network method was refined by using the water samples over stations such that the accuracy in chlorophyll-a estimation was increased. MERIS images are used in this study. Based on statistical analysis, RMSE value before and after the refinement is decreased from 6,7872 mg m-3 to 6,5606 mg m-3.
Normalisasi Radiometrik Relatif Multi Sensor dan Multi Temporal Berbasis Ekstraksi Fitur Pseudo-invariant Denaro, Lino Garda; Aldila Syariz, Muhammad; Nabilah, Salwa; Heriza, Dewinta; Lin, Chao-Hung
Jurnal Penginderaan Jauh Indonesia Vol 1 No 1 (2019)
Publisher : Masyarakat Ahli Penginderaan Jauh Indonesia (MAPIN) /Indonesian Society of Remote Sensing (ISRS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (683.449 KB)

Abstract

Relative Radiometric Normalization for Multi-sensors and Multi-temporal based on Pseudo-invariant Feature Extraction Technology developments promote the abundance availability of multi-temporal satellite imagery data. Otherwise, the utilization of such satellite data have not been implemented optimally due to several limitation requirements. The utilization of the multi-temporal satellite images for change detection is advantageous for modelling and predicting spatial changes in particular period of time. However, in the implementation, the radiometric correction on either multi-temporal or also multi-sensors is required. In this research, weighted regularized generalized canonical correlation analysis (WRGCCA) is proposed to select invariant features or pseudo-invariant features (PIFs) for multi-sensors and multi-temporal images. The method is the improvement of multivariate alteration detection (MAD) that adopts canonical correlation analysis (CCA) and its extension, generalized canonical correlation analysis (GCCA), to detect bitemporal and multi-temporal data respectively. However, each of the methods, CCA and GCCA, has the limitation on differentiating acceptable PIFs due to sensitive to spatial changes. Therefore, with the utilization of weighting and regularization functions to the algorithm, the proposed method WRGCCA with the aid of iterative reweighted multivariate alteration detection (IRMAD) can select reliable PIFs and yielding more accurate PIFs significantly to 10% - 50% determined from root mean square error (RMSE). The improved PIFs selection can be explained by qualitative and quantitative analysis based on the multi-temporal image normalization.
WorldView-2 Satellite Image Classification using U-Net Deep Learning Model Ilyas Ilyas; Lalu Muhamad Jaelani; Muhammad Aldila Syariz; Husnul Hidayat
Journal of Applied Geospatial Information Vol 5 No 2 (2021): Journal of Applied Geospatial Information (JAGI)
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jagi.v5i2.3150

Abstract

Land cover maps are important documents for local governments to perform urban planning and management. A field survey using measuring instruments can produce an accurate land cover map. However, this method is time-consuming, expensive, and labor-intensive. A number of researchers have proposed using remote sensing, which generates land cover maps using an optical satellite image with various statistical classification procedures. Recently, artificial intelligence (AI) technology, such as deep learning, has been used in multiple fields, including satellite image classification, with satisfactory results. In this study, a WorldView-2 image of Terangun in Aceh Province, which was acquired on Aug 2, 2016, was classified using a commonly used deep-learning-based classification, namely, U-net. There were eight classes used in the experiment: building, road, open land (such as green open space, bare land, grass, or low vegetation), river, farm, field, aquaculture pond, and garden. For comparison, three classification methods: maximum-likelihood, random forest, and support vector machine, were performed compared to U-Net. A land cover map provided by the government was used as a reference to evaluate the accuracy of land cover maps generated using two classification methods. The results with 100 randomly selected pixels revealed that U-Net was able to obtain a 72% and 0.585 for overall and kappa accuracy, respectively; whereas, overall accuracy and kappa accuracy for the maximum likelihood, random forest and support vector machine methods were 49% and 0.148; 59% and 0.392; and 67% and 0. 511; respectively. Therefore, U-Net outperformed those three of classification methods in classifying the image.
Co-Authors Denaro, Lino Garda Handoko, Eko Yuli Heriza, Dewinta Hidayat, Husnul Ilyas Ilyas Jaelani, Lalu Muhamad Lin, Chao-Hung Nabilah, Salwa Pratama Janur Wenda