Traditional Image Segmentation Research Articles

Research on tumors segmentation based on image enhancement method

One of the most effective ways to treat liver cancer is to perform precise liver resection surgery, the key step of which includes precise digital image segmentation of the liver and its tumor. However, traditional liver parenchymal segmentation techniques often face several challenges in performing liver segmentation: lack of precision, slow processing speed, and computational burden. These shortcomings limit the efficiency of surgical planning and execution. In this work, the model initially describes in detail a new image enhancement algorithm that enhances the key features of an image by adaptively adjusting the contrast and brightness of the image. Then, a deep learning-based segmentation network was introduced, which was specially trained on the enhanced images to optimize the detection accuracy of tumor regions. In addition, multi-scale analysis techniques have been incorporated into the study, allowing the model to analyze images at different resolutions to capture more nuanced tumor features. In the presentation of the experimental results, the study used the 3Dircadb dataset to test the effectiveness of the proposed method. The experimental results show that compared with the traditional image segmentation method, the new method using image enhancement technology has significantly improved the accuracy and recall rate of tumor identification.

Cross-Modality Medical Image Segmentation via Enhanced Feature Alignment and Cross Pseudo Supervision Learning.

Given the diversity of medical images, traditional image segmentation models face the issue of domain shift. Unsupervised domain adaptation (UDA) methods have emerged as a pivotal strategy for cross modality analysis. These methods typically utilize generative adversarial networks (GANs) for both image-level and feature-level domain adaptation through the transformation and reconstruction of images, assuming the features between domains are well-aligned. However, this assumption falters with significant gaps between different medical image modalities, such as MRI and CT. These gaps hinder the effective training of segmentation networks with cross-modality images and can lead to misleading training guidance and instability. To address these challenges, this paper introduces a novel approach comprising a cross-modality feature alignment sub-network and a cross pseudo supervised dual-stream segmentation sub-network. These components work together to bridge domain discrepancies more effectively and ensure a stable training environment. The feature alignment sub-network is designed for the bidirectional alignment of features between the source and target domains, incorporating a self-attention module to aid in learning structurally consistent and relevant information. The segmentation sub-network leverages an enhanced cross-pseudo-supervised loss to harmonize the output of the two segmentation networks, assessing pseudo-distances between domains to improve the pseudo-label quality and thus enhancing the overall learning efficiency of the framework. This method's success is demonstrated by notable advancements in segmentation precision across target domains for abdomen and brain tasks.

Significance of AI-assisted techniques for epiphyte plant monitoring and identification from drone images

Monitoring forest canopies is vital for ecological studies, particularly for assessing epiphytes in rain forest ecosystems. Traditional methods for studying epiphytes, such as climbing trees and building observation structures, are labor, cost intensive and risky. Unmanned Aerial Vehicles (UAVs) have emerged as a valuable tool in this domain, offering botanists a safer and more cost-effective means to collect data. This study leverages AI-assisted techniques to enhance the identification and mapping of epiphytes using UAV imagery. The primary objective of this research is to evaluate the effectiveness of AI-assisted methods compared to traditional approaches in segmenting/identifying epiphytes from UAV images collected in a reserve forest in Costa Rica. Specifically, the study investigates whether Deep Learning (DL) models can accurately identify epiphytes during complex backgrounds, even with a limited dataset of varying image quality. Systematically, this study compares three traditional image segmentation methods Auto Cluster, Watershed, and Level Set with two DL-based segmentation networks: the UNet and the Vision Transformer-based TransUNet. Results obtained from this study indicate that traditional methods struggle with the complexity of vegetation backgrounds and variability in target characteristics. Epiphyte identification results were quantitatively evaluated using the Jaccard score. Among traditional methods, Watershed scored 0.10, Auto Cluster 0.13, and Level Set failed to identify the target. In contrast, AI-assisted models performed better, with UNet scoring 0.60 and TransUNet 0.65. These results highlight the potential of DL approaches to improve the accuracy and efficiency of epiphyte identification and mapping, advancing ecological research and conservation.

Application of instance segmentation algorithm incorporating attention mechanism and BiFPN for sinter ore particle size recognition

The particle size composition of sinter plays a decisive role in the output and energy consumption of blast furnace ironmaking. To reduce energy consumption and achieve efficient production, accurate measurement of sinter particle size and distribution has become an important issue. Due to the problems of adhesion, large size difference and edge effect in sinter image, the traditional image segmentation algorithm makes it difficult to effectively separate sinter. This paper proposes an intelligent recognition method of sinter particle size based on an improved BlendMask instance segmentation algorithm. Bidirectional feature pyramid network is used to replace the feature pyramid network structure (FPN), and the attention mechanism Convolutional Block Attention Module (CBAM) is introduced to improve the ability to extract shallow features and enhance the ability to identify adherent sinter. The test results show that the average detection accuracy mean average precision (mAP) of the model proposed in this article is 71.0%, and the mAP50 is 93.8%. The average segmentation accuracy mAP is 68.0% and mAP50 is 93.8%. The identified sinter images were then edge-refined using morphological methods to collect statistics on the particle size distribution of the sinter. Compared with the actual measured sinter particle size, the average relative error is 4.9%. This method can accurately identify sinter in complex environments, improve the production efficiency of sinter, optimise resource utilisation and reduce personnel costs and has important application prospects.

Research on the PDC rock cuttings image instance segmentation method based on improved U-Net++ network and multi-feature fusion

Currently, PDC bits dominate the petroleum bit market. It is a small cut of polycrystalline diamond, which is embedded in the body of the drill bit. Due to the small number of cuttings generated through the PDC bit, the individual cuttings are small, and the cuttings edge is blurred, which results in poor effect and low precision of traditional image segmentation. Therefore, this paper proposes an image segmentation method regarding deep learning. Firstly, a multi-task learning method is introduced based on the U-Net segmentation model. A multi-task-learning-U-Net++ instance segmentation model is proposed. The results of semantic segmentation and edge segmentation are obtained by this segmentation model. Then, the cuttings are segmented by superpixel to obtain the sub-blocks of superpixel. Finally, a multi-feature fusion method is proposed, which integrates semantic information, edge information and superpixel sub-blocks to obtain the segmentation result of cuttings image. Experimental results show that the proposed method can effectively segment each cutting in the image and this method has certain robustness. Compared with the segmentation algorithms such as U-Net and U-Net++, this algorithm performs better in multiple image segmentation performance indexes.

Multi-strategy learning-based particle swarm optimization algorithm for COVID-19 threshold segmentation

With advancements in science and technology, the depth of human research on COVID-19 is increasing, making the investigation of medical images a focal point. Image segmentation, a crucial step preceding image processing, holds significance in the realm of medical image analysis. Traditional threshold image segmentation proves to be less efficient, posing challenges in selecting an appropriate threshold value. In response to these issues, this paper introduces Inner-based multi-strategy particle swarm optimization (IPSOsono) for conducting numerical experiments and enhancing threshold image segmentation in COVID-19 medical images. A novel dynamic oscillatory weight, derived from the PSO variant for single-objective numerical optimization (PSOsono) is incorporated. Simultaneously, the historical optimal positions of individuals in the particle swarm undergo random updates, diminishing the likelihood of algorithm stagnation and local optima. Moreover, an inner selection learning mechanism is proposed in the update of optimal positions, dynamically refining the global optimal solution. In the CEC 2013 benchmark test, PSOsono demonstrates a certain advantage in optimization capability compared to algorithms proposed in recent years, proving the effectiveness and feasibility of PSOsono. In the Minimum Cross Entropy threshold segmentation experiments for COVID-19, PSOsono exhibits a more prominent segmentation capability compared to other algorithms, showing good generalization across 6 CT images and further validating the practicality of the algorithm.

Enhance watershed algorithms using principal component analysis capabilities

Traditional image segmentation algorithms have many drawbacks, such as over-segmentation and image distortion due to reflected light. The Watershed algorithm is one of the most popular image segmentation algorithms. Over-segmentation errors caused by overlapping targets in the image, as well as noise and glare, must be removed. In this article, we apply image processing using the watershed algorithm and propose to improve the algorithm based on principal component analysis. PCA is a popular technique for analyzing large datasets with many advantages per observation. PCA improves data interpretability while maximizing information content, enabling visualization of multidimensional data by finding image component gradients in a new space called the principle component that is unaffected by noise and reflected light. In contrast, the components mainly containing noise will eliminate with negligible information. This paper introduces three primary steps. The process involves applying the watershed algorithm to the image in the first phase, using the proposed approach (applying the watershed algorithm and suggesting an improvement based on principal component analysis) to the image in the second step, and comparing the outcomes of the two previous processes. Test results show that the suggested technique can achieve accurate and durable target shapes.

Segmentation analysis of UAV images based on Unet deep learning algorithm

The continuous development of UAV technology provides us with more and higher quality data, in which the application of UAV image segmentation technology can help us better understand and process these data. Traditional image segmentation methods can no longer meet the needs of UAV image segmentation, so researchers have begun to explore the application of deep learning methods in UAV image segmentation.U-Net, as a classical deep learning model, is also widely used in UAV image segmentation.U-Net is characterized by two parts: encoder and decoder, which are used to extract the image features, and decoder is used to map the features back to the original image size. features to map back to the original image size. UAV image segmentation technology can be applied in agriculture, urban planning, environmental monitoring and other fields. In the field of agriculture, UAV image segmentation can help farmers better manage and monitor farmland to improve crop yield and quality. In the field of urban planning, UAV image segmentation can help urban planners better understand the development status of the city and provide a more scientific basis for urban planning. In the field of environmental monitoring, UAV image segmentation can help us better monitor the changes in the natural environment and provide more effective means for environmental protection. By setting the ratio of training set, validation set and test set as 6:2:2 and performing 100 rounds of training, the U-Net model shows good results in UAV image segmentation. The loss of the model gradually stabilizes at 0.2979 and the accuracy gradually converges to 90.34%. The test results show that the prediction results are very close to the true mask, indicating that the U-Net model can segment UAV images well. The application of UAV image segmentation technology can help us better understand and process the data acquired by UAV and provide more information and basis. In the future, with the continuous development of UAV technology, UAV image segmentation technology will also be more widely used. Through the application of UAV image segmentation technology, we will better understand and protect our environment, manage our farmland more efficiently, and plan our cities more scientifically.

Adaptive mask-based brain extraction method for head CT images.

Brain extraction is an important prerequisite for the automated diagnosis of intracranial lesions and determines, to a certain extent, the accuracy of subsequent lesion identification, localization, and segmentation. To address the problem that the current traditional image segmentation methods are fast in extraction but poor in robustness, while the Full Convolutional Neural Network (FCN) is robust and accurate but relatively slow in extraction, this paper proposes an adaptive mask-based brain extraction method, namely AMBBEM, to achieve brain extraction better. The method first uses threshold segmentation, median filtering, and closed operations for segmentation, generates a mask for the first time, then combines the ResNet50 model, region growing algorithm, and image properties analysis to further segment the mask, and finally complete brain extraction by multiplying the original image and the mask. The algorithm was tested on 22 test sets containing different lesions, and the results showed MPA = 0.9963, MIoU = 0.9924, and MBF = 0.9914, which were equivalent to the extraction effect of the Deeplabv3+ model. However, the method can complete brain extraction of approximately 6.16 head CT images in 1 second, much faster than Deeplabv3+, U-net, and SegNet models. In summary, this method can achieve accurate brain extraction from head CT images more quickly, creating good conditions for subsequent brain volume measurement and feature extraction of intracranial lesions.

Automatic segmentation of lamellar eutectoids in iron oxide scales using optimized U-net network

The variation in lamellar spacing of eutectoids in iron oxide affects the surface quality of the hot-rolled strip and its subsequent deep processing. In order to investigate the lamellar spacing numerically, an accurate segmentation of Fe and Fe3O4 phases is required in the structural picture. In this paper, a dataset was generated for identifying Fe and Fe3O4 in eutectoids. Also, a squeeze-and-excitation network module was integrated for improving the accuracy in image segmentation. The obtained experimental results show that the PA and mIoU could reach 96.83% and 94.49% in optimized U-net, indicating its excellent segmentation capability compared with traditional image segmentation networks. Using this method, fast and accurate extraction of the characteristic information about the picture can be realized, which would help in investigating the phase transition behavior of the eutectoid transformation in the Fe–O system.

Lightweight land cover classification via semantic segmentation of remote sensing imagery and analysis of influencing factors

Land cover classification is of great value and can be widely used in many fields. Earlier land cover classification methods used traditional image segmentation techniques, which cannot fully and comprehensively extract the ground information in remote sensing images. Therefore, it is necessary to integrate the advanced techniques of deep learning into the study of semantic segmentation of remote sensing images. However, most of current high-resolution image segmentation networks have disadvantages such as large parameters and high network training cost. In view of the problems above, a lightweight land cover classification model via semantic segmentation, DeepGDLE, is proposed in this paper. The model DeepGDLE is designed on the basis of DeeplabV3+ network and utilizes the GhostNet network instead of the backbone feature extraction network in the encoder. Using Depthwise Separable Convolution (DSC) instead of dilation convolution. This reduces the number of parameters and increases the computational speed of the model. By optimizing the dilation rate of parallel convolution in the ASPP module, the “grid effect” is avoided. ECANet lightweight channel attention mechanism is added after the feature extraction module and the pyramid pooling module to focus on the important weights of the model. Finally, the loss function Focal Loss is utilized to solve the problem of category imbalance in the dataset. As a result, the model DeepGDLE effectively reduces the parameters of the network model and the network training cost. And extensive experiments compared with several existing semantic segmentation algorithms such as DeeplabV3+, UNet, SegNet, etc. show that DeepGDLE improves the quality and efficiency of image segmentation. Therefore, compared to other networks, the DeepGDLE network model can be more effectively applied to land cover classification. In addition, in order to investigate the effects of different factors on the semantic segmentation performance of remote sensing images and to verify the robustness of the DeepGDLE model, a new remote sensing image dataset, FRSID, is constructed in this paper. This dataset takes into account more influences than the public dataset. The experimental results show that on the WHDLD dataset, the experimental metrics mIoU, mPA, and mRecall of the proposed model, DeepGDLE, are 62.29%, 72.85%, and 72.46%, respectively. On the FRSID dataset, the metrics mIoU, mPA, and mRecall are 65.89%, 74.43%, and 74.08%, respectively. For the future scope of research in this field, it may focus on the fusion of multi-source remote sensing data and the intelligent interpretation of remote sensing images.

Research on medical image segmentation technology based on deep learning

With economy expanding quickly, people's demand for medical services has become higher and higher, and medical image as an important basis for medical diagnosis has naturally received widespread attention. However, traditional image segmentation methods are easily affected by noise and unable to meet the complex and changing practical clinical applications. The increasing utilization of deep learning technology enables effective resolution of these problems. In this paper, we will first introduce the traditional image segmentation techniques, and describe the main methods to realize traditional image segmentation and its limitations. Immediately after that, it is proposed that deep learning methods can solve the challenges of medical image segmentation with traditional methods, and then the structure, algorithms and applications of several of the most commonly used deep learning methods are introduced. This paper proposes that medical image segmentation based on deep learning can segment the image more robustly and with high accuracy, and it can automatically obtain the most suitable features. The research in this paper will be of great value to the research and application of medical image segmentation technology based on deep learning.

Detection of Benggang in Remote Sensing Imagery through Integration of Segmentation Anything Model with Object-Based Classification

Benggang is a type of erosion landform that commonly occurs in the southern regions of China, posing significant threats to local farmland and human safety. Object-based classification (OBC) can be applied with high-resolution (HR) remote sensing images for detecting Benggang areas on a large spatial scale, offering essential data for aiding in the remediation efforts for these areas. Nevertheless, traditional image segmentation methods may face challenges in accurately delineating Benggang areas. Consequently, the extraction of spatial and textural features from these areas can be susceptible to inaccuracies, potentially compromising the detection accuracy of Benggang areas. To address this issue, this study proposed a novel approach that integrates Segment Anything Model (SAM) and OBC for Benggang detection. The SAM was used to segment HR remote sensing imagery to delineate the boundaries of Benggang areas. After that, the OBC was employed to identify Benggang areas based on spectral, geometrical, and textural features. In comparison to traditional pixel-based classification using the random forest classifier (RFC-PBC) and OBC based on the multi-resolution segmentation (MRS-OBC), the proposed SAM-OBC exhibited superior performance, achieving a detection accuracy of 85.46%, a false alarm rate of 2.19%, and an overall accuracy of 96.48%. The feature importance analysis conducted with random forests highlighted the GLDV Entropy, GLDV Angular Second Moment (ASM), and GLCM ASM as the most pivotal features for the identification of Benggang areas. Due to its inability to extract and utilize these textural features, the PBC yielded suboptimal results compared to both the SAM-OBC and MRS-OBC. In contrast to the MRS, the SAM demonstrated superior capabilities in the precise delineation of Benggang areas, ensuring the extraction of accurate textural and spatial features. As a result, the SAM-OBC significantly enhanced detection accuracy by 34.12% and reduced the false alarm rate by 2.06% compared to the MRS-OBC. The results indicate that the SAM-OBC performs well in Benggang detection, holding significant implications for the monitoring and remediation of Benggang areas.

SG-MIAN: Self-guided multiple information aggregation network for image-level weakly supervised skin lesion segmentation

Nowadays, skin disease is becoming one of the most malignant diseases that threaten people’s health. Computer aided diagnosis based on deep learning has become a widely used technology to assist medical professionals in diagnosis, and segmentation of lesion areas is one of the most important steps in it. However, traditional medical image segmentation methods rely on numerous pixel-level labels for fully supervised training, and such labeling process is time-consuming and requires professional competence. In order to reduce the costs of pixel-level labeling, we proposed a method only using image-level label to segment skin lesion areas. Due to the lack of lesion’s spatial and intensity information in image-level labels, and the wide distribution range of irregular shape and different texture on skin lesions, the algorithm must pay great attention to the automatic lesion localization and perception of lesion boundary. In this paper, we proposed a Self-Guided Multiple Information Aggregation Network (SG-MIAN). Our backbone network MIAN utilizes the Multiple Spatial Perceptron (MSP) solely using classification information as guidance to discriminate the key classification features of lesion areas, and thereby performing more accurate localization and activation of lesion areas. Additionally, adjunct to MSP, we also proposed an Auxiliary Activation Structure (AAS) and two auxiliary loss functions to further self-guided boundary correction, achieving the goal of accurate boundary activation. To verify the effectiveness of the proposed method, we conducted extensive experiments using the HAM10000 dataset and the PH2dataset, which demonstrated superior performance compared to most existing weakly supervised segmentation methods.

A multiangle polarised imaging-based method for thin sectionsegmentation.

The most crucial task of petroleum geology is to explore oil and gas reservoirs in the deep underground. As one of the analysis techniques in petroleum geological research, rock thin sectionidentification method includes particle segmentation, which is one of the key steps. A conventional sandstone thin sectionimage typically contains hundreds of mineral particles with blurred boundaries and complex microstructures inside the particles. Moreover, the complex lithology and low porosity of tight sandstone make traditional image segmentation methods unsuitable for solving the complex thin sectionsegmentation problems. This paper combines petrology and image processing technologies. First, polarised sequence images are aligned, and then the images are transformed to the HSV colour space to extract pores. Second, particles are extracted according to their extinction characteristics. Last, a concavity and corner detection matching method is used to process the extracted particles, thereby completing the segmentation of sandstone thin sectionimages. The experimental results show that our proposed method can more accurately fit the boundaries of mineral particles in sandstone images than existing image segmentation methods. Additionally, when applied in actual production scenarios, our method exhibits excellent performance, greatly improving thin sectionidentification efficiency and significantly assisting experts inidentification.

Attention‐based prohibited item detection in X‐ray images during security checking

AbstractThis paper focuses on the intelligent detection of prohibited items in X‐ray images during the security checking process. An intelligent semantic segmentation model of prohibited items in X‐ray images is proposed based on the attention‐based object localization method. Based on the pre‐trained CNN classification framework, the attention mechanism can map the high‐layer semantic information of objects into the input space, while generating energy saliency maps to locate the prohibited items. In order to make the obtained attention maps discriminative, the lateral and contrastive inhibition strategies are introduced and combined together which can highlight the responses of activated neurons. Under the guidance of attention responses, two traditional image segmentation algorithms are employed to achieve the semantic segmentation results for the prohibited items detection in X‐ray images. The proposed semantic segmentation model relies on weakly supervised learning mechanism, and only depends on the category labels of prohibited items, which greatly avoids the work cost of data semantic annotation. The experimental results based on the public SIXray baseline and the self‐built X‐ray image database demonstrate the proposed method can achieve about 65% IoU localization precise averagely. In addition, comparison experiments were carried out with the state‐of‐the‐arts and ablation experiments to verify the effectiveness of the proposed model.

Time-Series Field Phenotyping of Soybean Growth Analysis by Combining Multimodal Deep Learning and Dynamic Modeling.

The rate of soybean canopy establishment largely determines photoperiodic sensitivity, subsequently influencing yield potential. However, assessing the rate of soybean canopy development in large-scale field breeding trials is both laborious and time-consuming. High-throughput phenotyping methods based on unmanned aerial vehicle (UAV) systems can be used to monitor and quantitatively describe the development of soybean canopies for different genotypes. In this study, high-resolution and time-series raw data from field soybean populations were collected using UAVs. The RGB (red, green, and blue) and infrared images are used as inputs to construct the multimodal image segmentation model-the RGB & Infrared Feature Fusion Segmentation Network (RIFSeg-Net). Subsequently, the segment anything model was employed to extract complete individual leaves from the segmentation results obtained from RIFSeg-Net. These leaf aspect ratios facilitated the accurate categorization of soybean populations into 2 distinct varieties: oval leaf type variety and lanceolate leaf type variety. Finally, dynamic modeling was conducted to identify 5 phenotypic traits associated with the canopy development rate that differed substantially among the classified soybean varieties. The results showed that the developed multimodal image segmentation model RIFSeg-Net for extracting soybean canopy cover from UAV images outperformed traditional deep learning image segmentation networks (precision = 0.94, recall = 0.93, F1-score = 0.93). The proposed method has high practical value in the field of germplasm resource identification. This approach could lead to the use of a practical tool for further genotypic differentiation analysis and the selection of target genes.

MIS-Net: A deep learning-based multi-class segmentation model for CT images.

The accuracy of traditional CT image segmentation algorithms is hindered by issues such as low contrast and high noise in the images. While numerous scholars have introduced deep learning-based CT image segmentation algorithms, they still face challenges, particularly in achieving high edge accuracy and addressing pixel classification errors. To tackle these issues, this study proposes the MIS-Net (Medical Images Segment Net) model, a deep learning-based approach. The MIS-Net model incorporates multi-scale atrous convolution into the encoding and decoding structure with symmetry, enabling the comprehensive extraction of multi-scale features from CT images. This enhancement aims to improve the accuracy of lung and liver edge segmentation. In the evaluation using the COVID-19 CT Lung and Infection Segmentation dataset, the left and right lung segmentation results demonstrate that MIS-Net achieves a Dice Similarity Coefficient (DSC) of 97.61. Similarly, in the Liver Tumor Segmentation Challenge 2017 public dataset, the DSC of MIS-Net reaches 98.78.

Pancreatic cancer pathology image segmentation with channel and spatial long-range dependencies

Based on deep learning, pancreatic cancer pathology image segmentation technology effectively assists pathologists in achieving improved treatment outcomes. However, compared to traditional image segmentation tasks, the large size of tissues in pathology images requires a larger receptive field. While methods based on dilated convolutions or attention mechanisms can enhance the receptive field, they cannot capture long-range feature dependencies. Directly applying self-attention mechanisms to capture long-range dependencies results in intolerable computational complexity. To address these challenges, we introduce a channel and spatial self-attention (CS) Module designed for efficiently capturing both channel and spatial long-range feature dependencies in pancreatic cancer pathological images. Specifically, the channel and spatial self-attention module consists of an adaptive channel self-attention module and a window-shift spatial self-attention module. The adaptive channel self-attention module adaptively pools features to a fixed size to capture long-range feature dependencies. While the window-shift spatial self-attention module captures spatial long-range dependencies in a window-based manner. Additionally, we propose a re-weighted cross-entropy loss to mitigate the impact of long-tail distribution on performance. Our proposed method surpasses state-of-the-art on both our Pancreatic Cancer Pathology Image (PCPI) dataset and the GlaS challenge dataset. The mDice and mIoU have achieved 73.93% and 59.42% in our PCPI dataset.

Multidimensional characterization of particle morphology and mineralogical composition using CT data and R-vine copulas

Computed tomography (CT) can capture volumes large enough to measure a statistically meaningful number of micron-sized particles with a sufficiently good resolution to allow for the analysis of individual particles. However, the development of methods to efficiently investigate such image data and interpretably model the observed particle features is still an active field of research. When image data of particles exhibiting a wide range of shapes and sizes is considered, traditional image segmentation methods, such as the classic watershed algorithm, struggle to recognize particles with satisfying accuracy. Thus, more advanced methods of machine learning must be utilized for image segmentation to improve the validity of subsequent analyses. Moreover, CT data does not include information about the mineralogical composition of particles and, therefore, additional SEM-EDS image data has to be acquired. In this paper, micro-CT image data of a particle system mostly consisting of zinnwaldite–quartz composites is considered. First, an image segmentation method is applied which uses deep convolutional neural networks, in particular an adaptation of the U-net architecture. This has the advantage of requiring less hand-labeling than other machine learning methods, while also being more flexible with the possibility of transfer learning. Then, fully parameterized models based on vine copulas are designed to determine multivariate probability distributions of descriptor vectors for the size, shape, texture and composition of particles—allowing for the estimation and interpretable characterization of interdependencies between particle descriptors. For model fitting, the segmented three-dimensional CT data and co-registered two-dimensional SEM-EDS data are used. The models are applied to predict the mineralogical composition of particles, solely on the basis of particle descriptors observed in CT data.