Multispectral Image Segmentation Research Articles

Early detection of Botrytis cinerea symptoms using deep learning multi-spectral image segmentation

Botrytis cinerea is an airborne plant pathogen that causes grey mould disease on horticultural crops, such as tomato, cucumber, pepper, and more. Early detection is crucial for timely and effective strategies for crop loss prevention. In this study, in planta assays were set up enabling a thorough assessment of grey mould progress on cucumber plants both artificially inoculated with two different inoculation methods, and naturally infected throughout the whole growing season of the host under controlled conditions simulating the environment typically prevailing in greenhouses. Multi-spectral imaging was used to capture a novel dataset across various spectral bands containing multiple stages of infection. Deep learning (DL) segmentation architectures, combined with Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) encoders, were employed to identify B. cinerea at various infection stages. The U-Net++ model with a MobileViT-S encoder performed best, achieving a Dice Coefficient (DC) of 0.677, an Intersection over Union (IoU) of 0.656, and a recall rate of 0.807, with a high overall accuracy of 90.1 %. In regards to early detection, the model achieves an IoU of 0.375 on 2nd day post inoculation (dpi), 0.230 on 1st dpi and 0.437 on the 6th dpi. The infection stages were compared to similar visible symptoms of physiological decline or plant responses to abiotic stress. The outcomes of this study demonstrate an important advance in the Artificial Intelligence (AI) plant health detection, for early crop disease diagnosis.

Correction to: Multispectral image segmentation utilizing constrained clustering approach and CGT classifier

Correction to: Multispectral image segmentation utilizing constrained clustering approach and CGT classifier

Multispectral image segmentation utilizing constrained clustering approach and CGT classifier

Multispectral image segmentation utilizing constrained clustering approach and CGT classifier

Dynamic monitoring of surface area and water volume of reservoirs using satellite imagery, computer vision and deep learning

The development of surface area and estimated volume measurements for water reservoirs using advanced computational techniques has become an increasingly important task for reservoir operation. Thus, the objective of this study is to determine the surface area and estimate the volume of water reservoirs through the segmentation of multispectral satellite images using a Fully Convolutional Neural Network (FCN). This paper introduces a universally applicable framework that utilizes deep convolutional neural networks for segmenting water bodies, particularly focusing on areas in Paraíba State, Northeast Brazil, susceptible to severe droughts. The study utilized Sentinel-2 satellite data obtained from Google Earth Engine for the period between January 2019 and September 2023. The extraction of water surface area from satellite images was achieved using the U-Net model, a deep learning framework based on FCN. The accuracy of water surface area extraction was analyzed using the Intersection-Over-Union (IoU) metric, also known as Jaccard Index. Subsequently, reservoir volumes were calculated using elevation-area-volume curves. The evaluation metrics for the results included root mean square error (RMSE), mean absolute percentage error (MAPE), coefficient of determination (R2), and Pearson correlation coefficient (r). The network demonstrated high accuracy, achieving a 95% IoU on the test set. This methodology was applied to four reservoirs, yielding promising results that closely aligned the monitored reservoir curves derived from satellite imagery with observed field data. The findings of this study are significant for research on semiarid region changes, extending their relevance to a broader understanding of global environmental dynamics. This work not only contributes to the field of remote sensing in hydrology but also offers practical insights for effective water resource management in drought-prone regions.

Brain tumor segmentation of the FLAIR MRI images using novel ResUnet

New technologies are growing faster and now play a key role in analysing new ways of looking at the morphology of the brain. It is difficult to diagnose a brain tumor. Accurate diagnosis and segmentation of the brain tumor used for early treatment planning. Different neuroimaging modalities are used to provide better tissue resolution and provide assistance to the radiologist. Manual segmentation of the brain tumor is a complicated task as it faces problems with noise, intensity inhomogeneity, merging of tissues, and overlapping of tissue intensity. This makes manual segmentation a time consuming approach. In recent CAD systems are developed using deep learning models. In this paper, we used residual models and form as a Unet to perform segmentation of the tissues. To perform segmentation, we used the Kaggle LGG dataset, which contains 110 patient datasets. We designed a novel residual modelused as the backbone to design Unet from scratch and perform segmentation of multi spectral images. The Proposed model works well, and the performance of the model is analysed using the Dice coefficient, Jaccard index/IoU. The model yields a dice coefficient of 0.9056 and a jaccard index/ IoU of 0.8293. This model gives better tissue segmentation using FLAIR modality compared to the existing frame work.

FAPNET: Feature Fusion with Adaptive Patch for Flood-Water Detection and Monitoring.

In satellite remote sensing applications, waterbody segmentation plays an essential role in mapping and monitoring the dynamics of surface water. Satellite image segmentation-examining a relevant sensor data spectrum and identifying the regions of interests to obtain improved performance-is a fundamental step in satellite data analytics. Satellite image segmentation is challenging for a number of reasons, which include cloud interference, inadequate label data, low lighting and the presence of terrain. In recent years, Convolutional Neural Networks (CNNs), combined with (satellite captured) multispectral image segmentation techniques, have led to promising advances in related research. However, ensuring sufficient image resolution, maintaining class balance to achieve prediction quality and reducing the computational overhead of the deep neural architecture are still open to research due to the sophisticated CNN hierarchical architectures. To address these issues, we propose a number of methods: a multi-channel Data-Fusion Module (DFM), Neural Adaptive Patch (NAP) augmentation algorithm and re-weight class balancing (implemented in our PHR-CB experimental setup). We integrated these techniques into our novel Fusion Adaptive Patch Network (FAPNET). Our dataset is the Sentinel-1 SAR microwave signal, used in the Microsoft Artificial Intelligence for Earth competition, so that we can compare our results with the top scores in the competition. In order to validate our approach, we designed four experimental setups and in each setup, we compared our results with the popular image segmentation models UNET, VNET, DNCNN, UNET++, U2NET, ATTUNET, FPN and LINKNET. The comparisons demonstrate that our PHR-CB setup, with class balance, generates the best performance for all models in general and our FAPNET approach outperforms relative works. FAPNET successfully detected the salient features from the satellite images. FAPNET with a MeanIoU score of 87.06% outperforms the state-of-the-art UNET, which has a score of 79.54%. In addition, FAPNET has a shorter training time than other models, comparable to that of UNET (6.77 min for 5 epochs). Qualitative analysis also reveals that our FAPNET model successfully distinguishes micro waterbodies better than existing models. FAPNET is more robust to low lighting, cloud and weather fluctuations and can also be used in RGB images. Our proposed method is lightweight, computationally inexpensive, robust and simple to deploy in industrial applications. Our research findings show that flood-water mapping is more accurate when using SAR signals than RGB images. Our FAPNET architecture, having less parameters than UNET, can distinguish micro waterbodies accurately with shorter training time.

Land Cover Segmentation of Multispectral Images Using Multiresolution Algorithm

Land Cover Segmentation of Multispectral Images Using Multiresolution Algorithm

Acceleration of the Advanced Segmentation Algorithm for Multispectral Images Using CNN

Proposed an improved approach to the segmentation of multispectral images using Convolutional Neural Networks (CNN). The original algorithm was described earlier. It took into account some errors that could arise during the processing of SNA images using a sliding window. The proposed modification uses the and indices, which have a high correlation coefficient with real objects present in the images, also images pyramids were used.

Semantic Segmentation of Multispectral Images via Linear Compression of Bands: An Experiment Using RIT-18

Semantic segmentation of remotely sensed imagery is a basic task for many applications, such as forest monitoring, cloud detection, and land-use planning. Many state-of-the-art networks used for this task are based on RGB image datasets and, as such, prefer three-band images as their input data. However, many remotely sensed images contain more than three spectral bands. Although it is technically possible to feed multispectral images directly to those networks, poor segmentation accuracy was often obtained. To overcome this issue, the current image dimension reduction methods are either to use feature extraction or to select an optimal combination of three bands through different trial processes. However, it is well understood that the former is often comparatively less effective, because it is not optimized towards segmentation accuracy, while the latter is less efficient due to repeated trial selections of three bands for the optimal combination. Therefore, it is meaningful to explore alternative methods that can utilize multiple spectral bands efficiently in the state-of-the-art networks for semantic segmentation of similar accuracy as the trial selection approach. In this study, a hot-swappable stem structure (LC-Net) is proposed to linearly compress the input bands to fit the input preference of typical networks. For the three commonly used network structures tested on the RIT-18 dataset (having six spectral bands), the approach proposed was found to be an equivalently effective but much more efficient alternative to the trial selection approach.

Semantic segmentation of multispectral photoacoustic images using deep learning

Photoacoustic (PA) imaging has the potential to revolutionize functional medical imaging in healthcare due to the valuable information on tissue physiology contained in multispectral photoacoustic measurements. Clinical translation of the technology requires conversion of the high-dimensional acquired data into clinically relevant and interpretable information. In this work, we present a deep learning-based approach to semantic segmentation of multispectral photoacoustic images to facilitate image interpretability. Manually annotated photoacoustic and ultrasound imaging data are used as reference and enable the training of a deep learning-based segmentation algorithm in a supervised manner. Based on a validation study with experimentally acquired data from 16 healthy human volunteers, we show that automatic tissue segmentation can be used to create powerful analyses and visualizations of multispectral photoacoustic images. Due to the intuitive representation of high-dimensional information, such a preprocessing algorithm could be a valuable means to facilitate the clinical translation of photoacoustic imaging.

Variational-Scale Segmentation for Multispectral Remote-Sensing Images Using Spectral Indices

Many studies have focused on performing variational-scale segmentation to represent various geographical objects in high-resolution remote-sensing images. However, it remains a significant challenge to select the most appropriate scales based on the geographical-distribution characteristics of ground objects. In this study, we propose a variational-scale multispectral remote-sensing image segmentation method using spectral indices. Real scenes in remote-sensing images contain different types of land cover with different scales. Therefore, it is difficult to segment images optimally based on the scales of different ground objects. To guarantee image segmentation of ground objects with their own scale information, spectral indices that can be used to enhance some types of land cover, such as green cover and water bodies, were introduced into marker generation for the watershed transformation. First, a vector field model was used to determine the gradient of a multispectral remote-sensing image, and a marker was generated from the gradient. Second, appropriate spectral indices were selected, and the kernel density estimation was used to generate spectral-index marker images based on the analysis of spectral indices. Third, a series of mathematical morphology operations were used to obtain a combined marker image from the gradient and the spectral index markers. Finally, the watershed transformation was used for image segmentation. In a segmentation experiment, an optimal threshold for the spectral-index-marker generation method was identified. Additionally, the influence of the scale parameter was analyzed in a segmentation experiment based on a five-subset dataset. The comparative results for the proposed method, the commonly used watershed segmentation method, and the multiresolution segmentation method demonstrate that the proposed method yielded multispectral remote-sensing images with much better performance than the other methods.

Segmentation of aerospace images using convolutional neural networks

To solve practical problems, including for the management of territories, monitoring of emergencies, for the development of the ecology of our planet, for the study and ecological research of our planet, effective algorithms for the segmentation of multispectral images are needed. In recent years, image segmentation using convolutional neural networks has become very popular. A significant difference of this approach is the annotation of the assignment of each pixel to a particular class of objects, so that the learning process of such networks is completely controlled. The paper proposes a new method for segmentation of aerospace im-ages of high spatial resolution based on convolutional neural networks and mask generation. Our model is based on a combined U-Net network with MobileNetV2 as the backbone. It trains on ground data and provides a full prediction mask. A network has been implemented, consisting of separate networks of the same class, working on the expansion of segmentation. Semantic features are used to reduce errors at the semantic level of the outline. Since an in-dependent test revealed some shortcomings in certain classes of the earth's surface, further re-search will be devoted to the creation of a catalog of training samples to support successful differentiation of objects. In general, the obtained accuracy estimates demonstrate the modern characteristics of the developed model, as well as the effectiveness of this combination of network and datasets for test regions. The results show that the proposed algorithm can effec-tively improve the overall accuracy of the semantic segmentation of high spatial resolution remote sensing images and reduce the training time and segmentation time.

The Influence of Seasonality on the Multi-Spectral Image Segmentation for Identification of Abandoned Land

Areas of agricultural land in Lithuania have decreased from 2005 to 2021 by up to 2.4%. Agricultural lands that are no longer used for their main purpose are very likely to become abandoned and the emergence of such lands can cause a variety of social, economic, and environmental problems. Therefore, it is very important to constantly monitor changes of abandoned agricultural lands. The purpose of the research is to analyse the influence of seasonality on image segmentation for the identification of abandoned land areas. Multi-spectral Sentinel-2 images from different periods (April, July, and September) and three supervised image segmentation methods (Spectral Angle Mapping (SAM), Maximum_Likelihood (ML), and Minimum distance (MD)) were used with the same parameters in this research. Studies had found that the most appropriate time to segment abandoned lands was in September, according to the SAM and ML algorithms. During this period, the intensity of the green colour was the highest and the colour brightness of abandoned lands differed from the colour intensity of other lands.

Maximum 3D Tsallis entropy based multilevel thresholding of brain MR image using attacking Manta Ray foraging optimization

Nevertheless, the accuracy of a multilevel image thresholding technique using 1D or 2D Tsallis entropy is limited. To overcome this, we propose a maximum 3D Tsallis entropy-based multilevel thresholding method. The idea of 3D Tsallis entropy is introduced. Opposed to the 1D/2D Tsallis entropy, the 3D Tsallis entropy-based approach is more robust, it performs well even in the case of the low signal-to-noise-ratio and contrast. Manta Ray Foraging Optimization (MRFO) algorithm is a newly introduced algorithm to solve the optimization problem by imitating the foraging technique of Manta Ray fish in the ocean using a mathematical model. Due to insufficient energy levels of search agents in MRFO, they fail to avoid local minima and fall on it. To make the algorithm more effective for the segmentation application, we introduce a new algorithm coined as attacking Manta Ray foraging optimization (AMRFO). A set of classical benchmark functions together with composite functions (CEC 2014) is used to validate the proposed AMRFO algorithm. Statistical analysis is implicitly carried out using Wilcoxon’s signed-rank test and Friedman’s mean rank test. Interestingly, the results show that the proposed AMRFO is superior to the state-of-the-art optimization algorithms. Moreover, the proposed method is also compared with 1D/2D Tsallis entropy-based approaches. To experiment, 100 test images from the AANLIB MR Image dataset are considered. Our method outperforms 1D/2D Tsallis entropy-based approaches. The proposed scheme would be useful for the segmentation of multi-spectral color images.

Detection of Potential Vernal Pools on the Canadian Shield (Ontario) Using Object-Based Image Analysis in Combination with Machine Learning

Vernal pools are small, temporary, forested wetlands of ecological importance with a high sensitivity to changing climate and land-use patterns. These ecosystems are under considerable development pressure in southeastern Georgian Bay, where mapping techniques are required to inform wise land-use decisions. Our mapping approach combines common machine learning techniques [random forest, support vector machines (SVMs)] with object-based image analysis. Using multispectral image segmentation on high-resolution orthoimagery, we first created objects and assigned classes based on field collected data. We then supplied machine learning algorithms with data from freely available sources (Ontario orthoimagery and Sentinel 2) and tested accuracy on a reserved dataset. We achieved producer’s accuracies of 85 and 79% and user’s accuracies of 78 and 84% for random forest and SVMs models, respectively. Difficulty differentiating between small, dark shadows and small, obscured pools accounted for many of the omission and commission errors. Our automated approach of vernal pool classification provides a relatively accurate, consistent, and fast mapping strategy compared to manual photointerpretation. Our models can be applied on a regional basis to help verify the locations of pools in an area of Ontario that is in critical need of more detailed ecological information.

Semantic segmentation of multispectral satellite images for land use analysis based on embedded information

Semantic segmentation of satellite and aerial imageries has many applications including automated map making, land use analysis, urban planning and so on. This paper presents a special type of semantic segmentation. First, the boundaries of natural objects such as agricultural fields are detected and approximated by polygons. Second, texture recognition based on Digital Wavelet Transform (DWT) and Local Binary Patterns (LBPs) is implemented using a limited textural dictionary. The obtained information is embedded using DWT and Arnold’s transform. Such watermarked images can be publicly available, but semantic information after extraction and inpainting is provided only to the authorized users. Such semantic labelling is very useful in land use analysis of rural territories.

MAPPING EXTENSION AND MAGNITUDE OF CHANGES INDUCED BY CYCLONE IDAI WITH MULTI-TEMPORAL LANDSAT AND SAR IMAGES

Abstract. In this paper it is described a study case of a rapid assessment of change detections for post-cyclone Idai vegetated damage and flood extension estimation by fusion of multi-temporal Landsat and sentinel-1 SAR images. For automated change detection, after disasters, many algorithms have been proposed. To visualize the changes induced by cyclone we tested and compared two automated change detection techniques namely: Principal Components Analysis (PCA), Normalized Difference Vegetation Index (NDVI) and image segmentation. With the image segmentation of multispectral and SAR images, it was possible to visualize the extension of the wet area. For this specific application, PCA was identified as the optimal change detection indicator than NDVI. This study suggested that image segmentation, principal components analysis, and normalized difference vegetation index can be used for change detection of surface water due to flood and disasters especially in prone countries like Mozambique.

Interface design for monitoring and estimation system for flooding through an image analysis of remote sensing (SAVUI)

In recent years the floods in Mexico caused economic and human losses, therefore, it is necessary to use the possible tools that can provide help to the government to reduce damage from natural disasters. For this, we decided to develop a graphical user interface, known as GUI in Matlab for the segmentation of SAR, Multispectral and POLSAR images, with the intention of detecting flooding and vulnerable areas to flooding. The designed software compute a rivers segmentation in order to make the comparison between image with flooding and the image without flooding from the same area, and to obtain a visually result where a projection of the vulnerable areas to flooding in the original image this with help of basic segmentation algorithms such as grayscale, binarization, dilation, wavelet, normalization, filtering and edge detection.

Multi-Spectral Image Segmentation Based on the K-means Clustering

Agriculture is one of the oldest economic aspects of human civilisation, and it is still undergoing a dynamic makeover in the course of the application of IT innovative mechanisms in farming methodology. Remote sensing has vied a significant role in crop classification, crop health and yield assessment. Multispectral remote sensing plays a vital role in providing enhancement of more detailed analysis of crop segmentation. In this article, pixel-based clustering of 12 channels is carried out using the satellite image from Sentinel 2 remote sensing satellite via k-means clustering. K-means clustering algorithm is usually a better method of classifying high-resolution satellite imagery. The extracted regions are classified using a minimum distance decision rule.

Multispectral Image Segmentation Based on a Fuzzy Clustering Algorithm Combined with Tsallis Entropy and a Gaussian Mixture Model

Accurate multispectral image segmentation is essential in remote sensing research. Traditional fuzzy clustering algorithms used to segment multispectral images have several disadvantages, including: (1) they usually only consider the pixels’ grayscale information and ignore the interaction between pixels; and, (2) they are sensitive to noise and outliers. To overcome these constraints, this study proposes a multispectral image segmentation algorithm based on fuzzy clustering combined with the Tsallis entropy and Gaussian mixture model. The algorithm uses the fuzzy Tsallis entropy as regularization item for fuzzy C-means (FCM) and improves dissimilarity measure using the negative logarithm of the Gaussian Mixture Model (GMM). The Hidden Markov Random Field (HMRF) is introduced to define prior probability of neighborhood relationship, which is used as weights of the Gaussian components. The Lagrange multiplier method is used to solve the segmentation model. To evaluate the proposed segmentation algorithm, simulated and real multispectral images were segmented using the proposed algorithm and two other algorithms for comparison (i.e., Tsallis Fuzzy C-means (TFCM), Kullback–Leibler Gaussian Fuzzy C-means (KLG-FCM)). The study found that the modified algorithm can accelerate the convergence speed, reduce the effect of noise and outliers, and accurately segment simulated images with small gray level differences with an overall accuracy of more than 98.2%. Therefore, the algorithm can be used as a feasible and effective alternative in multispectral image segmentation, particularly for those with small color differences.