Concealed Object Detection

PSTNet: Enhanced Polyp Segmentation With Multi-Scale Alignment and Frequency Domain Integration

Accurate segmentation of colorectal polyps in colonoscopy images is crucial for effective diagnosis and management of colorectal cancer (CRC). However, current deep learning-based methods primarily rely on fusing RGB information across multiple scales, leading to limitations in accurately identifying polyps due to restricted RGB domain information and challenges in feature misalignment during multi-scale aggregation. To address these limitations, we propose the Polyp Segmentation Network with Shunted Transformer (PSTNet), a novel approach that integrates both RGB and frequency domain cues present in the images. PSTNet comprises three key modules: the Frequency Characterization Attention Module (FCAM) for extracting frequency cues and capturing polyp characteristics, the Feature Supplementary Alignment Module (FSAM) for aligning semantic information and reducing misalignment noise, and the Cross Perception localization Module (CPM) for synergizing frequency cues with high-level semantics to achieve efficient polyp segmentation. Extensive experiments on challenging datasets demonstrate PSTNet's significant improvement in polyp segmentation accuracy across various metrics, consistently outperforming state-of-the-art methods. The integration of frequency domain cues and the novel architectural design of PSTNet contribute to advancing computer-assisted polyp segmentation, facilitating more accurate diagnosis and management of CRC.

UGEE-Net: Uncertainty-guided and edge-enhanced network for image splicing localization

Image splicing, a prevalent method for image tampering, has significantly undermined image authenticity. Existing methods for Image Splicing Localization (ISL) struggle with challenges like limited accuracy and subpar performance when dealing with imperceptible tampering and multiple tampered regions. We introduce an Uncertainty-Guided and Edge-Enhanced Network (UGEE-Net) for ISL to tackle these issues. UGEE-Net consists of two core tasks: uncertainty guidance and edge enhancement. We employ Bayesian learning to model uncertainty maps of tampered regions, directing the model's focus to challenging pixels. Simultaneously, we employ a frequency domain-auxiliary edge enhancement strategy to imbue localization features with global contour information and fine-grained local details. These mechanisms work in parallel, synergistically boosting performance. Additionally, we introduce a cross-level fusion and propagation mechanism that effectively utilizes contextual information for cross-layer feature integration and leverages channel-level correlations for cross-layer feature propagation, gradually enhancing the localization feature's details. Experiment results affirm UGEE-Net's superiority in terms of detection accuracy, robustness, and generalization capabilities. Furthermore, to meet the growing demand for high-quality datasets in image forensics, we present the HTSI12K dataset, which includes 12,000 spliced images with imperceptible tampering traces and diverse categories, rendering it suitable for real-world auxiliary model training.

WoodenCube: An Innovative Dataset for Object Detection in Concealed Industrial Environments

With the rapid advancement of intelligent manufacturing technologies, the operating environments of modern robotic arms are becoming increasingly complex. In addition to the diversity of objects, there is often a high degree of similarity between the foreground and the background. Although traditional RGB-based object-detection models have achieved remarkable success in many fields, they still face the challenge of effectively detecting targets with textures similar to the background. To address this issue, we introduce the WoodenCube dataset, which contains over 5000 images of 10 different types of blocks. All images are densely annotated with object-level categories, bounding boxes, and rotation angles. Additionally, a new evaluation metric, Cube-mAP, is proposed to more accurately assess the detection performance of cube-like objects. In addition, we have developed a simple, yet effective, framework for WoodenCube, termed CS-SKNet, which captures strong texture features in the scene by enlarging the network's receptive field. The experimental results indicate that our CS-SKNet achieves the best performance on the WoodenCube dataset, as evaluated by the Cube-mAP metric. We further evaluate the CS-SKNet on the challenging DOTAv1.0 dataset, with the consistent enhancement demonstrating its strong generalization capability.

Validating polyp and instrument segmentation methods in colonoscopy through Medico 2020 and MedAI 2021 Challenges

Automatic analysis of colonoscopy images has been an active field of research motivated by the importance of early detection of precancerous polyps. However, detecting polyps during the live examination can be challenging due to various factors such as variation of skills and experience among the endoscopists, lack of attentiveness, and fatigue leading to a high polyp miss-rate. Therefore, there is a need for an automated system that can flag missed polyps during the examination and improve patient care. Deep learning has emerged as a promising solution to this challenge as it can assist endoscopists in detecting and classifying overlooked polyps and abnormalities in real time, improving the accuracy of diagnosis and enhancing treatment. In addition to the algorithm's accuracy, transparency and interpretability are crucial to explaining the whys and hows of the algorithm's prediction. Further, conclusions based on incorrect decisions may be fatal, especially in medicine. Despite these pitfalls, most algorithms are developed in private data, closed source, or proprietary software, and methods lack reproducibility. Therefore, to promote the development of efficient and transparent methods, we have organized the "Medico automatic polyp segmentation (Medico 2020)" and "MedAI: Transparency in Medical Image Segmentation (MedAI 2021)" competitions. The Medico 2020 challenge received submissions from 17 teams, while the MedAI 2021 challenge also gathered submissions from another 17 distinct teams in the following year. We present a comprehensive summary and analyze each contribution, highlight the strength of the best-performing methods, and discuss the possibility of clinical translations of such methods into the clinic. Our analysis revealed that the participants improved dice coefficient metrics from 0.8607 in 2020 to 0.8993 in 2021 despite adding diverse and challenging frames (containing irregular, smaller, sessile, or flat polyps), which are frequently missed during a routine clinical examination. For the instrument segmentation task, the best team obtained a mean Intersection over union metric of 0.9364. For the transparency task, a multi-disciplinary team, including expert gastroenterologists, accessed each submission and evaluated the team based on open-source practices, failure case analysis, ablation studies, usability and understandability of evaluations to gain a deeper understanding of the models' credibility for clinical deployment. The best team obtained a final transparency score of 21 out of 25. Through the comprehensive analysis of the challenge, we not only highlight the advancements in polyp and surgical instrument segmentation but also encourage subjective evaluation for building more transparent and understandable AI-based colonoscopy systems. Moreover, we discuss the need for multi-center and out-of-distribution testing to address the current limitations of the methods to reduce the cancer burden and improve patient care.

A frequency selection network for medical image segmentation

Existing medical image segmentation methods may only consider feature extraction and information processing in spatial domain, or lack the design of interaction between frequency information and spatial information, or ignore the semantic gaps between shallow and deep features, and lead to inaccurate segmentation results. Therefore, in this paper, we propose a novel frequency selection segmentation network (FSSN), which achieves more accurate lesion segmentation by fusing local spatial features and global frequency information, better design of feature interactions, and suppressing low correlation frequency components for mitigating semantic gaps. Firstly, we propose a global-local feature aggregation module (GLAM) to simultaneously capture multi-scale local features in the spatial domain and exploits global frequency information in the frequency domain, and achieves complementary fusion of local details features and global frequency information. Secondly, we propose a feature filter module (FFM) to mitigate semantic gaps when we conduct cross-level features fusion, and makes FSSN discriminatively determine which frequency information should be preserved for accurate lesion segmentation. Finally, in order to make better use of local information, especially the boundary of lesion region, we employ deformable convolution (DC) to extract pertinent features in the local range, and makes our FSSN can focus on relevant image contents better. Extensive experiments on two public benchmark datasets show that compared with representative medical image segmentation methods, our FSSN can obtain more accurate lesion segmentation results in terms of both objective evaluation indicators and subjective visual effects with fewer parameters and lower computational complexity.

Enhancing tunnel crack detection with linear seam using mixed stride convolution and attention mechanism

Cracks in tunnel lining structures constitute a common and serious problem that jeopardizes the safety of traffic and the durability of the tunnel. The similarity between lining seams and cracks in terms of strength and morphological characteristics renders the detection of cracks in tunnel lining structures challenging. To address this issue, a new deep learning-based method for crack detection in tunnel lining structures is proposed. First, an improved attention mechanism is introduced for the morphological features of lining seams, which not only aggregates global spatial information but also features along two dimensions, height and width, to mine more long-distance feature information. Furthermore, a mixed strip convolution module leveraging four different directions of strip convolution is proposed. This module captures remote contextual information from various angles to avoid interference from background pixels. To evaluate the proposed approach, the two modules are integrated into a U-shaped network, and experiments are conducted on Tunnel200, a tunnel lining crack dataset, as well as the publicly available crack datasets Crack500 and DeepCrack. The results show that the approach outperforms existing methods and achieves superior performance on these datasets.

Multimodal fusion for anticipating human decision performance

Anticipating human decisions while performing complex tasks remains a formidable challenge. This study proposes a multimodal machine-learning approach that leverages image features and electroencephalography (EEG) data to predict human response correctness in a demanding visual searching task. Notably, we extract a novel set of image features pertaining to object relationships using the Segment Anything Model (SAM), which enhances prediction accuracy compared to traditional features. Additionally, our approach effectively utilizes a combination of EEG signals and image features to streamline the feature set required for the Random Forest Classifier (RFC) while maintaining high accuracy. The findings of this research hold substantial potential for developing advanced fault alert systems, particularly in critical decision-making environments such as the medical and defence sectors.

Boundary-Sensitive Segmentation of Small Liver Lesions

Early diagnosis plays a pivotal role in handling the global health challenge posed by liver diseases. However, early-stage lesions are typically quite small, presenting significant difficulties due to insufficient regions for developing effective features, indistinguishable boundaries of small lesions, and a lack of tiny liver lesion masks. To address these issues, we approach the solution in two-fold: an efficient model and a high-quality dataset. The model is built upon the advantages of path signature and camouflaged object detection. The path signature narrows down the ambiguous boundaries between lesions and other tissues while the camouflaged object detection achieves high accuracy in detecting inconspicuous lesions. The two are seamlessly integrated to ensure high accuracy and fidelity. For the dataset, we collect more than ten thousand liver images with over four thousand lesions, approximately half of which are small. Experiments on both an established dataset and our newly constructed one show that the proposed model outperforms state-of-the-art semantic segmentation and camouflaged object detection models, particularly in detecting small lesions. Moreover, the decisive and faithful salience maps generated by the model at the boundary regions demonstrate its strong robustness.

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.

Features Split and Aggregation Network for Camouflaged Object Detection

Higher standards have been proposed for detection systems since camouflaged objects are not distinct enough, making it possible to ignore the difference between their background and foreground. In this paper, we present a new framework for Camouflaged Object Detection (COD) named FSANet, which consists mainly of three operations: spatial detail mining (SDM), cross-scale feature combination (CFC), and hierarchical feature aggregation decoder (HFAD). The framework simulates the three-stage detection process of the human visual mechanism when observing a camouflaged scene. Specifically, we have extracted five feature layers using the backbone and divided them into two parts with the second layer as the boundary. The SDM module simulates the human cursory inspection of the camouflaged objects to gather spatial details (such as edge, texture, etc.) and fuses the features to create a cursory impression. The CFC module is used to observe high-level features from various viewing angles and extracts the same features by thoroughly filtering features of various levels. We also design side-join multiplication in the CFC module to avoid detail distortion and use feature element-wise multiplication to filter out noise. Finally, we construct an HFAD module to deeply mine effective features from these two stages, direct the fusion of low-level features using high-level semantic knowledge, and improve the camouflage map using hierarchical cascade technology. Compared to the nineteen deep-learning-based methods in terms of seven widely used metrics, our proposed framework has clear advantages on four public COD datasets, demonstrating the effectiveness and superiority of our model.

Zero-Shot Camouflaged Object Detection

The goal of Camouflaged object detection (COD) is to detect objects that are visually embedded in their surroundings. Existing COD methods only focus on detecting camouflaged objects from seen classes, while they suffer from performance degradation to detect unseen classes. However, in a real-world scenario, collecting sufficient data for seen classes is extremely difficult and labeling them requires high professional skills, thereby making these COD methods not applicable. In this paper, we propose a new zero-shot COD framework (termed as ZSCOD), which can effectively detect the never unseen classes. Specifically, our framework includes a Dynamic Graph Searching Network (DGSNet) and a Camouflaged Visual Reasoning Generator (CVRG). In details, DGSNet is proposed to adaptively capture more edge details for boosting the COD performance. CVRG is utilized to produce pseudo-features that are closer to the real features of the seen camouflaged objects, which can transfer knowledge from seen classes to unseen classes to help detect unseen objects. Besides, our graph reasoning is built on a dynamic searching strategy, which can pay more attention to the boundaries of objects for reducing the influences of background. More importantly, we construct the first zero-shot COD benchmark based on the COD10K dataset. Experimental results on public datasets show that our ZSCOD not only detects the camouflaged object of unseen classes but also achieves state-of-the-art performance in detecting seen classes.

Early Detection of Human Decision-Making in Concealed Object Visual Searching Tasks: An EEG-BiLSTM Study

Detecting concealed objects presents a significant challenge for human and artificial intelligent systems. Detecting concealed objects task necessitates a high level of human attention and cognitive effort to complete the task successfully. Thus, in this study, we use concealed objects as stimuli for our decision-making experimental paradigms to quantify participants' decision-making performance. We applied a deep learning model, Bi-directional Long Short Term Memory (BiLSTM), to predict the participant's decision accuracy by using their electroencephalogram (EEG) signals as input. The classifier model demonstrated high accuracy, reaching 96.1% with an epoching time range of 500 ms following the stimulus event onset. The results revealed that the parietal-occipital brain region provides highly informative information for the classifier in the concealed visual searching tasks. Furthermore, the neural mechanism underlying the concealed visual-searching and decision-making process was explained by analyzing serial EEG components. The findings of this study could contribute to the development of a fault alert system, which has the potential to improve human decision-making performance.

Edge-Guided Camouflaged Object Detection via Multi-Level Feature Integration

Camouflaged object detection (COD) aims to segment those camouflaged objects that blend perfectly into their surroundings. Due to the low boundary contrast between camouflaged objects and their surroundings, their detection poses a significant challenge. Despite the numerous excellent camouflaged object detection methods developed in recent years, issues such as boundary refinement and multi-level feature extraction and fusion still need further exploration. In this paper, we propose a novel multi-level feature integration network (MFNet) for camouflaged object detection. Firstly, we design an edge guidance module (EGM) to improve the COD performance by providing additional boundary semantic information by combining high-level semantic information and low-level spatial details to model the edges of camouflaged objects. Additionally, we propose a multi-level feature integration module (MFIM), which leverages the fine local information of low-level features and the rich global information of high-level features in adjacent three-level features to provide a supplementary feature representation for the current-level features, effectively integrating the full context semantic information. Finally, we propose a context aggregation refinement module (CARM) to efficiently aggregate and refine the cross-level features to obtain clear prediction maps. Our extensive experiments on three benchmark datasets show that the MFNet model is an effective COD model and outperforms other state-of-the-art models in all four evaluation metrics (Sα, Eϕ, Fβw, and MAE).

FSNet: Focus Scanning Network for Camouflaged Object Detection

Camouflaged object detection (COD) aims to discover objects that blend in with the background due to similar colors or textures, etc. Existing deep learning methods do not systematically illustrate the key tasks in COD, which seriously hinders the improvement of its performance. In this paper, we introduce the concept of focus areas that represent some regions containing discernable colors or textures, and develop a two-stage focus scanning network for camouflaged object detection. Specifically, a novel encoder-decoder module is first designed to determine a region where the focus areas may appear. In this process, a multi-layer Swin transformer is deployed to encode global context information between the object and the background, and a novel cross-connection decoder is proposed to fuse cross-layer textures or semantics. Then, we utilize the multi-scale dilated convolution to obtain discriminative features with different scales in focus areas. Meanwhile, the dynamic difficulty aware loss is designed to guide the network paying more attention to structural details. Extensive experimental results on the benchmarks, including CAMO, CHAMELEON, COD10K, and NC4K, illustrate that the proposed method performs favorably against other state-of-the-art methods.

MGL: Mutual Graph Learning for Camouflaged Object Detection

Camouflaged object detection, which aims to detect/segment the object(s) that blend in with their surrounding, remains challenging for deep models due to the intrinsic similarities between foreground objects and background surroundings. Ideally, an effective model should be capable of finding valuable clues from the given scene and integrating them into a joint learning framework to co-enhance the representation. Inspired by this observation, we propose a novel Mutual Graph Learning (MGL) model by shifting the conventional perspective of mutual learning from regular grids to graph domain. Specifically, an image is decoupled by MGL into two task-specific feature maps - one for finding the rough location of the target and the other for capturing its accurate boundary details. Then, the mutual benefits can be fully exploited by reasoning their high-order relations through graphs recurrently. It should be noted that our method is different from most mutual learning models that model all between-task interactions with the use of a shared function. To increase information interactions, MGL is built with typed functions for dealing with different complementary relations. To overcome the accuracy loss caused by interpolation to higher resolution and the computational redundancy resulting from recurrent learning, the S-MGL is equipped with a multi-source attention contextual recovery module, called R-MGL_v2, which uses the pixel feature information iteratively. Experiments on challenging datasets, including CHAMELEON, CAMO, COD10K, and NC4K demonstrate the effectiveness of our MGL with superior performance to existing state-of-the-art methods. The code can be found at https://github.com/fanyang587/MGL.

Salient Objects in Clutter

In this paper, we identify and address a serious design bias of existing salient object detection (SOD) datasets, which unrealistically assume that each image should contain at least one clear and uncluttered salient object. This design bias has led to a saturation in performance for state-of-the-art SOD models when evaluated on existing datasets. However, these models are still far from satisfactory when applied to real-world scenes. Based on our analyses, we propose a new high-quality dataset and update the previous saliency benchmark. Specifically, our dataset, called Salient Objects in Clutter (SOC), includes images with both salient and non-salient objects from several common object categories. In addition to object category annotations, each salient image is accompanied by attributes that reflect common challenges in common scenes, which can help provide deeper insight into the SOD problem. Further, with a given saliency encoder, e.g., the backbone network, existing saliency models are designed to achieve mapping from the training image set to the training ground-truth set. We therefore argue that improving the dataset can yield higher performance gains than focusing only on the decoder design. With this in mind, we investigate several dataset-enhancement strategies, including label smoothing to implicitly emphasize salient boundaries, random image augmentation to adapt saliency models to various scenarios, and self-supervised learning as a regularization strategy to learn from small datasets. Our extensive results demonstrate the effectiveness of these tricks. We also provide a comprehensive benchmark for SOD, which can be found in our repository: https://github.com/DengPingFan/SODBenchmark.

Deep Gradient Learning for Efficient Camouflaged Object Detection

This paper introduces deep gradient network (DGNet), a novel deep framework that exploits object gradient supervision for camouflaged object detection (COD). It decouples the task into two connected branches, i.e., a context and a texture encoder. The essential connection is the gradient-induced transition, representing a soft grouping between context and texture features. Benefiting from the simple but efficient framework, DGNet outperforms existing state-of-the-art COD models by a large margin. Notably, our efficient version, DGNet-S, runs in real-time (80 fps) and achieves comparable results to the cutting-edge model JCSOD-CVPR21 with only 6.82% parameters. The application results also show that the proposed DGNet performs well in the polyp segmentation, defect detection, and transparent object segmentation tasks. The code will be made available at https://github.com/GewelsJI/DGNet.

CAVER: Cross-Modal View-Mixed Transformer for Bi-Modal Salient Object Detection

Most of the existing bi-modal (RGB-D and RGB-T) salient object detection methods utilize the convolution operation and construct complex interweave fusion structures to achieve cross-modal information integration. The inherent local connectivity of the convolution operation constrains the performance of the convolution-based methods to a ceiling. In this work, we rethink these tasks from the perspective of global information alignment and transformation. Specifically, the proposed cross-modal view-mixed transformer (CAVER) cascades several cross-modal integration units to construct a top-down transformer-based information propagation path. CAVER treats the multi-scale and multi-modal feature integration as a sequence-to-sequence context propagation and update process built on a novel view-mixed attention mechanism. Besides, considering the quadratic complexity w.r.t. the number of input tokens, we design a parameter-free patch-wise token re-embedding strategy to simplify operations. Extensive experimental results on RGB-D and RGB-T SOD datasets demonstrate that such a simple two-stream encoder-decoder framework can surpass recent state-of-the-art methods when it is equipped with the proposed components.

MAGNet: A Camouflaged Object Detection Network Simulating the Observation Effect of a Magnifier

In recent years, protecting important objects by simulating animal camouflage has been widely employed in many fields. Therefore, camouflaged object detection (COD) technology has emerged. COD is more difficult to achieve than traditional object detection techniques due to the high degree of fusion of objects camouflaged with the background. In this paper, we strive to more accurately and efficiently identify camouflaged objects. Inspired by the use of magnifiers to search for hidden objects in pictures, we propose a COD network that simulates the observation effect of a magnifier called the MAGnifier Network (MAGNet). Specifically, our MAGNet contains two parallel modules: the ergodic magnification module (EMM) and the attention focus module (AFM). The EMM is designed to mimic the process of a magnifier enlarging an image, and AFM is used to simulate the observation process in which human attention is highly focused on a particular region. The two sets of output camouflaged object maps were merged to simulate the observation of an object by a magnifier. In addition, a weighted key point area perception loss function, which is more applicable to COD, was designed based on two modules to give greater attention to the camouflaged object. Extensive experiments demonstrate that compared with 19 cutting-edge detection models, MAGNet can achieve the best comprehensive effect on eight evaluation metrics in the public COD dataset. Additionally, compared to other COD methods, MAGNet has lower computational complexity and faster segmentation. We also validated the model's generalization ability on a military camouflaged object dataset constructed in-house. Finally, we experimentally explored some extended applications of COD.

Polarization-driven camouflaged object segmentation via gated fusion

Recently, polarization-based models for camouflaged object segmentation have attracted research attention. However, to construct this camouflaged object segmentation model, the main challenge is to effectively fuse polarization and light intensity features. Therefore, we propose a multi-modal camouflaged object segmentation method via gated fusion. First, the spatial positioning module is designed to perform channel calibration and global spatial attention alignment between polarization mode and light intensity mode from high-level feature representation to locate object positioning accurately. Then, the gated fusion module (GFM) is designed to selectively fuse the object information contained in the polarization and light intensity features. Among them, semantic information of location features is introduced in the GFM to guide each mode to aggregate dominant features. Finally, the features of each layer are aggregated to obtain an accurate segmentation result map. At the same time, considering the lack of public evaluation and training data on light intensity-polarization (I-P) camouflaged detection, we build the light I-P camouflaged detection dataset. Experimental results demonstrate that our proposed method outperforms other typical multi-modal segmentation methods in this dataset.

Selecting Post-Processing Schemes for Accurate Detection of Small Objects in Low-Resolution Wide-Area Aerial Imagery

In low-resolution wide-area aerial imagery, object detection algorithms are categorized as feature extraction and machine learning approaches, where the former often requires a post-processing scheme to reduce false detections and the latter demands multi-stage learning followed by post-processing. In this paper, we present an approach on how to select post-processing schemes for aerial object detection. We evaluated combinations of each of ten vehicle detection algorithms with any of seven post-processing schemes, where the best three schemes for each algorithm were determined using average F-score metric. The performance improvement is quantified using basic information retrieval metrics as well as the classification of events, activities and relationships (CLEAR) metrics. We also implemented a two-stage learning algorithm using a hundred-layer densely connected convolutional neural network for small object detection and evaluated its degree of improvement when combined with the various post-processing schemes. The highest average F-scores after post-processing are 0.902, 0.704 and 0.891 for the Tucson, Phoenix and online VEDAI datasets, respectively. The combined results prove that our enhanced three-stage post-processing scheme achieves a mean average precision (mAP) of 63.9% for feature extraction methods and 82.8% for the machine learning approach.