Traditional Image Segmentation Algorithms Research Articles

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.

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.

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.

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.

Unsupervised leaf segmentation in complex backgrounds using mutual information minimization

Leaf segmentation is crucial for plant recognition, especially for tree species identification. In natural environments, leaf segmentation can be very challenging due to the lack of prior information about leaves and the variability of backgrounds. In typical applications, supervised algorithms often require pixel-level annotation of regions, which can be labour-intensive and limited to identifying plant species using pre-labelled samples. On the other hand, traditional unsupervised image segmentation algorithms require specialised parameter tuning for leaf images to achieve optimal results. Therefore, this paper proposes an unsupervised leaf segmentation method that combines mutual information with neural networks to better generalise to unknown samples and adapt to variations in leaf shape and appearance to distinguish and identify different tree species. First, a model combining a Variational Autoencoder (VAE) and a segmentation network is used as a pre-segmenter to obtain dynamic masks. Secondly, the dynamic masks are combined with the segmentation masks generated by the mask generator module to construct the initial mask. Then, the patcher module uses the Mutual Information Minimum (MIM) loss as an optimisation objective to reconstruct independent regions based on this initial mask. The process of obtaining dynamic masks through pre-segmentation is unsupervised, and the entire experimental process does not involve any label information. The experimental method was performed on tree leaf images with a naturally complex background using the publicly available Pl@ntLeaves dataset. The results of the experiment showed that compared to existing excellent methods on this dataset, the IoU (Intersection over Union) index increased by 3.9%.

Bone tumor examination based on FCNN-4s and CRF fine segmentation fusion algorithm

Background and objectiveBone tumor is a kind of harmful orthopedic disease, there are benign and malignant points. Aiming at the problem that the accuracy of the existing machine learning algorithm for bone tumor image segmentation is not high, a bone tumor image segmentation algorithm based on improved full convolutional neural network which consists fully convolutional neural network (FCNN-4s) and conditional random field (CRF). MethodologyThe improved fully convolutional neural network (FCNN-4s) was used to perform coarse segmentation on preprocessed images. Batch normalization layers were added after each convolutional layer to accelerate the convergence speed of network training and improve the accuracy of the trained model. Then, a fully connected conditional random field (CRF) was fused to refine the bone tumor boundary in the coarse segmentation results, achieving the fine segmentation effect. ResultsThe experimental results show that compared with the traditional convolutional neural network bone tumor image segmentation algorithm, the algorithm has a great improvement in segmentation accuracy and stability, the average Dice can reach 91.56%, the real-time performance is better. ConclusionCompared with the traditional convolutional neural network segmentation algorithm, the algorithm in this paper has a more refined structure, which can effectively solve the problem of over-segmentation and under-segmentation of bone tumors. The segmentation prediction has better real-time performance, strong stability, and can achieve higher segmentation accuracy.

Interactive texture replacement of cartoon characters based on deep learning model

Abstract To understand the deep learning model, the author proposed the research of interactive texture replacement of cartoon characters. For image segmentation, if you want to fill a cartoon without any texture in detail, or replace the unsatisfied texture area, first, we need to separate the filled or replaced area from the cartoon. For this reason, the traditional image segmentation algorithm has been carefully studied and analyzed, and the author chooses the Graphcut texture synthesis algorithm, the algorithm is parallelized and improved, and the innovative point of lighting customization is proposed based on the original algorithm, which can affect the synthesis effect according to the input lighting image. In terms of timeliness and synthesis effect, the Graphcut algorithm has been improved. Experimental results show that the algorithm can maintain the brightness distribution of the original cartoon and the practicability and efficiency of the algorithm proposed by the author.

RETRACTED ARTICLE: Evaluation of image segmentation and multi class object recognition algorithm based on machine learning

This paper mainly updates the image segmentation and multi-target recognition algorithm through the support vector machine algorithm in ML, and verifies the reliability of the new proposed algorithm through simulation experiments, and determines that the ML-based image segmentation and multi-target recognition algorithm can have stronger processing capability in image recognition and processing. Finally, the performance of traditional image segmentation and multi class object recognition algorithms and the image segmentation and multi class object recognition algorithms proposed in this paper were compared through experiments. The results showed that the performance of the ML based image segmentation and multi class object recognition algorithms proposed in this paper has increased by 29.1% on average in all aspects.

Extraction of tree crown parameters of high-density pure Cunninghamia lanceolata plantations by combining the U-Net model and watershed algorithm.

As one of the important timber species in China, Cunninghamia lanceolata is widely distributed in southern China. The information of tree individuals and crown plays an important role in accurately monitoring forest resources. Therefore, it is particularly significant to accurately grasp such information of individual C. lanceolata tree. For high-canopy closed forest stands, the key to correctly extract such information is whether the crowns of mutual occlusion and adhesion can be accurately segmented. Taking the Fujian Jiangle State-owned Forest Farm as the research area and using the UAV image as the data source, we developed a method to extract crown information of individual tree based on deep learning method and watershed algorithm. Firstly, the deep learning neural network model U-Net was used to segment the coverage area of the canopy of C. lanceolata, and then the traditional image segmentation algorithm was used to segment the individual tree to obtain the number and crown information of individual tree. Under the condition of maintaining the same training set, validation set and test set, the extraction results of the canopy coverage area by the U-Net model and traditional machine learning methods [random forest (RF) and support vector machine (SVM)] were compared. Then, two individual tree segmentation results were compared, one using the marker-controlled watershed algorithm, and the other using the combination of the U-Net model and marker-controlled watershed algorithm. The results showed that the segmentation accuracy (SA), precision, IoU (intersection over union) and F1-score (harmonic mean of precision and recall) of the U-Net model were higher than those of RF and SVM. Compared with RF, the value of those four indicators increased by 4.6%, 14.9%, 7.6% and 0.05, respectively. Compared with SVM, the four indicators increased by 3.3%, 8.5%, 8.1% and 0.05, respectively. In terms of extracting the number of trees, the overall accuracy (OA) of the U-Net model combined with the marker-controlled watershed algorithm was 3.7% higher than that of the marker-controlled watershed algorithm, with the mean absolute error (MAE) being decreased by 3.1%. In terms of extracting crown area and crown width of individual tree, R2 increased by 0.11 and 0.09, mean squared error decreased by 8.49 m2 and 4.27 m, and MAE decreased by 2.93 m2 and 1.72 m, respectively. The combination of deep learning U-Net model and watershed algorithm could overcome the challenges in accurately extracting the number of trees and the crown information of individual tree of high-density pure C. lanceolata plantations. It was an efficient and low-cost method of extracting tree crown parameters, which could provide a basis for developing intelligent forest resource monitoring.

Intelligent analysis method of dam material gradation for asphalt-core rock-fill dam based on enhanced Cascade Mask R-CNN and GCNet

Reasonable dam materials’ gradation design for asphalt-core rock-fill dams is one of the main ways to control permeability. It is a challenge to test whether it can meet the requirements of dam construction. The computer vision method provides a new idea for asphalt-core rock-fill dam material gradation testing. However, due to the characteristics of densely overlapping and multi-scale sizes of dam material particles, the traditional image segmentation methods and algorithms cannot achieve accurate segmentation of dam materials’ images, and it is hard to apply the segmentation result to quantify the gradation curve. In this research, the enhanced Cascade Mask R-CNN with ResNet and PAFPN (Path Aggregation Feature Pyramid Networks) is proposed. Multi-scale features extracted by ResNet and feature ensemble can be realized using PAFPN. Data augmentation (DA) and online hard example mining (OHEM) are also applied in segmentation model training. Moreover, the GCNet is proposed to calibrate the gradation curve. The nonlinear relationship between the real gradation and the one based on the segmentation results can be revealed and the model of dam materials’ gradation analysis can be established. In the research, the enhanced Cascade Mask R-CNN can achieve 84.2 mAP, which is higher than that of Cascade Mask R-CNN with 74.9 mAP. The effectiveness of the proposed module and training strategies is proved using ablation experiments. The average error of each level for the gradation calibration using GCNet is 0.55%, 1.87%, 2.22%, 1.18%, and 2.42% respectively. The accuracy can meet the requirements of hydraulic engineering construction, which verifies the effectiveness of the GCNet network for gradation calibration, and the research provides a new method and technology for intelligent gradation testing of the asphalt-core rock-fill dam.

Image segmentation for blind lanes based on improved SegNet model

The position of blind lanes must be correctly determined in order for blind people to travel safely. Aiming at the low accuracy and slow speed of traditional blind lanes image segmentation algorithms, a semantic segmentation method based on SegNet and MobileNetV3 is proposed. The main idea is to replace the coding part of the original SegNet model with the feature extraction part of MobileNetV3 and remove the pooling layer. Blind lanes images were collected through online search and self-shooting, and then the data were manually marked by LabelMe software and trained on TensorFlow deep learning framework. The experimental results show that the improved model has high segmentation accuracy and recognition speed. The pixel accuracy of blind lanes segmentation is 98.21%, the mean intersection over union is 96.29%, and the average time for processing a 416 × 416 image is 0.057 s, which meets the real-time requirements of the blind guidance system.

A New Smoke Segmentation Method Based on Improved Adaptive Density Peak Clustering

Smoke image segmentation plays a vital role in the accuracy of target extraction. In order to improve the performance of the traditional fire image segmentation algorithm, a new smoke segmentation method based on improved double truncation distance self-adaptive density peak clustering(TSDPC) is proposed. Firstly, the smoke image is over-segmented into multiple superpixels to reduce the time cost, and the local density of sample points corresponding to each superpixel is redefined by location information and color space information. Secondly, TSDPC combines the information entropy theory to find the optimal double truncation distance. Finally, TSDPC uses trigonometric functions to determine clustering centers in the decision diagram, which can solve the problem of over-segmentation. Then, it assigns labels to the remain sample points for obtaining the clustering result. Compared with other algorithms, the accuracy of TSDPC is increased by 5.68% on average, and the F1 value is increased by 6.69% on average, which shows its high accuracy and effectiveness. In public dataset, TSDPC has also demonstrated its effectiveness.

Pattern segmentation based on PoolNet and boundary connectivity

Traditional image segmentation algorithms need to design features manually. Therefore, based on the advantage of PoolNet in object detection, this paper proposes a clothing pattern segmentation algorithm combining PoolNet detection and boundary connectivity. This algorithm has more fine edges than traditional segmentation. Especially in the case of similar background before and after, we can get more complete pattern elements, and have better results in segmentation accuracy.

ESA-UNet for assisted diagnosis of cardiac magnetic resonance image based on the semantic segmentation of the heart.

Cardiovascular diseases have become the number one disease affecting human health in today's society. In the diagnosis of cardiac diseases, magnetic resonance image (MRI) technology is the most widely used one. However, in clinical diagnosis, the analysis of MRI relies on manual work, which is laborious and time-consuming, and also easily influenced by the subjective experience of doctors. In this article, we propose an artificial intelligence-aided diagnosis system for cardiac MRI with image segmentation as the main component to assist in the diagnosis of cardiovascular diseases. We first performed adequate pre-processing of MRI. The pre-processing steps include the detection of regions of interest of cardiac MRI data, as well as data normalization and data enhancement, and then we input the images after data pre-processing into the deep learning network module of ESA-Unet for the identification of the aorta in order to obtain preliminary segmentation results, and finally, the boundaries of the segmentation results are further optimized using conditional random fields. For ROI detection, we first use standard deviation filters for filtering to find regions in the heart cycle image sequence where pixel intensity varies strongly with time and then use Canny edge detection and Hough transform techniques to find the region of interest containing the heart. The ESA-Unet proposed in this article, moreover, is jointly designed with a self-attentive mechanism and multi-scale jump connection based on convolutional networks. The experimental dataset used in this article is from the Department of CT/MRI at the Second Affiliated Hospital of Fujian Medical University. Experiments compare other convolution-based methods, such as UNet, FCN, FPN, and PSPNet, and the results show that our model achieves the best results on Acc, Pr, ReCall, DSC, and IoU metrics. After comparative analysis, the experimental results show that the ESA-UNet network segmentation model designed in this article has higher accuracy, intuitiveness, and more application value than traditional image segmentation algorithms. With the continuous application of nuclear magnetic resonance technology in clinical diagnosis, the method in this article is expected to become a tool that can effectively improve the efficiency of doctors' diagnoses.

Femoral image segmentation based on two-stage convolutional network using 3D-DMFNet and 3D-ResUnet

ObjectiveThe femur is a typical human long bone with an irregular spatial structure. Femoral fractures are the most common occurrence in middle-aged and older adults. The structure of human bone tissue is very complex, and there are significant differences between individuals. Segmenting bone tissue is a challenging task and of great practical significance. MethodsOur research is based on segmenting and the three-dimensional reconstruction of femoral images using X-ray imaging. The currently commonly used two-dimensional fully convolutional network Unet has the problem of ignoring spatial position information and losing too much feature information. The commonly used three-dimensional fully convolutional network 3D Unet has the problem of ignoring spatial position information and losing too much feature information. For the problem of many model parameters, we proposes a two-stage network segmentation model composed of 3D-DMFNet and 3D-ResUnet networks and trains the network in stages to segment the femur. One stage is used to detect the coarse segmentation of the femur range, and one stage is used for the fine segmentation of the femur so that the training speed is fast and the segmentation accuracy is moderate, which is suitable for detecting the femur range. ResultsThe experimental dataset used in this paper is from The Second Affiliated Hospital of Fujian Medical University, which consists of 30 sets of femur X-ray images. The experiment compares the accuracy and loss value of Unet and the two-stage convolutional network. The image shows that the two-stage convolutional network has higher accuracy. At the same time, this paper shows the effect of the two-stage coarse segmentation and fine segmentation of medical images. Subsequently, this paper applies the model to practice and obtains the model's Dice, Sensitivity, Specificity and Pixel Accuracy values. After comparative analysis, the experimental results show that the two-stage network segmentation model composed of 3D-DMFNet and 3D-ResUnet network designed in this paper has higher accuracy, intuitiveness, and more application value than traditional image segmentation algorithms. ConclusionWith the continuous application of X-ray images in clinical diagnosis using femoral images, the method in this paper is expected to become a diagnostic tool that can effectively improve the accuracy and loss of femoral image segmentation and the three-dimensional reconstruction.

Segmentation for Human Motion Injury Ultrasound Medical Images Using Deep Feature Fusion

Image processing technology assists physicians in the analysis of athletes’ human motion injuries, not only to improve the accuracy of athletes’ injury detection but also to improve the localization and recognition of injury locations. It is important to accurately segment human motion injury ultrasound medical images. To address many problems such as poor effect of traditional ultrasonic medical image segmentation algorithm for a sports injury. Therefore, we propose a segmentation algorithm for human motion injury ultrasound medical images using deep feature fusion. First, the accurate estimated value of human posture is extracted and combined with image texture features and image gray value as the target feature value of the ultrasonic medical image of human motion injury. Second, the image features are deeply fused by an adaptive fusion algorithm to enhance the image resolution. Finally, the best segmentation value of the image is obtained by the trained support vector machine to realize the accurate segmentation of human motion injury ultrasonic medical image. The results show that the average accuracy of the posture accurate estimation of the proposed algorithm is 95.97%; the segmentation time of the human motion injury ultrasound medical image of the proposed algorithm is below 150 ms; and the convergence of the algorithm is completed when the number of iterations is 3. The maximum segmentation error rate is 2.68%. The image segmentation effect is consistent with the ideal target segmentation effect. The proposed algorithm has important application value in the field of ultrasonic medical diagnosis of sports injury.

An Encoder-Decoder Network for Automatic Clinical Target Volume Target Segmentation of Cervical Cancer in CT Images

Cervical cancer is a common gynecological cancer, and its common treatment method radiotherapy depends on target area delineation. The manual delineation work takes a long time and has low accuracy, so automating such delineation is important. At present, some traditional image segmentation algorithms for target area delineation have low accuracy rates. Deep learning algorithms also face some difficulties, such as insufficient data and long training time. As the popular network used in medical image segmentation, U-net still has several disadvantages when handling small targets with unclear boundaries. According to the characteristics of the clinical target volume target segmentation task of cervical cancer, this study modified the U-net structure and optimized the training loss to improve the accuracy of small target detection. The modified structure could handle target boundaries well with operations such as bilinear upsampling. Finally, the proposed algorithm was evaluated on the dataset and compared with several deep learning-based algorithms. Results indicate that the proposed approach has certain superiority.

Artificial Intelligence-Based Semisupervised Self-Training Algorithm in Pathological Tissue Image Segmentation.

In the field of medical image processing, due to the differences in tissues, organs, and imaging methods, obtained medical images have significant differences. With the development of intelligence in medicine, an increasing number of computing optimization algorithms based on AI technology have also been applied to the field of medicine. Because the image segmentation algorithm based on the semisupervised self-training algorithm solves initialization class center large randomness problem in the traditional cluster-based image segmentation algorithm, this article aims to integrate the artificial intelligence semisupervised self-training algorithm into the pathological tissue image segmentation problem. An experimental group is designed to collect sample images and the algorithm proposed in this article is used to perform image segmentation to achieve a better visual experience and images. Although there is no general image segmentation theory, many scholars have been committed to applying new concepts and new methods to image segmentation in recent years and combining specific theoretical image segmentation methods has achieved good application results in image segmentation. For example, wavelet analysis, wavelet transform, neural networks, and genetic algorithms can effectively improve the segmentation effect. The results of the Seg cutting method designed in this article show that, in retinal blood vessel segmentation results on a database of healthy people, the sensitivity value is 0.941633, the false-positive rate is 0.952933, the specificity is 0.956787, and the accuracy rate is 0.96182, which are all higher than those in other methods. Image cutting methods such as FNN, CNN, and AWN have addressed the case tissue image cutting problem. Using the Seg cutting method designed in this article to segment the retinal blood vessels on a diabetes patient database, the sensitivity value is 0.8106, the false-positive rate is 0.0511, the specificity is 0.9712, the accuracy is 0.9421, and the false-positive rate is omitted. The false-positive rate is lower than AWN, and other indicators are higher than FNN, CNN, AWN, and other image cutting methods. The application of artificial intelligence-based semisupervised self-training algorithms in pathological tissue image segmentation is realized.

Extraction Method of Target Region for HIFU Therapy Based on Ultrasonic Image Semantic Segmentation

Accurate extraction of the therapeutic target region in the monitoring ultrasonic image is necessary for high intensity focused ultrasound (HIFU) therapy. However, due to the interference of complex background and large amount of noise in the ultrasonic image, traditional image segmentation algorithms have limitations on the target region segmentation of ultrasonic image. In order to extract the target region in ultrasonic image, a semantic segmentation method combining improved DeepLabv3+ network and improved loss function is proposed. On the basis of DeepLabv3+ network, dense atrous spatial pyramid pooling (DenseASPP) enhanced by attention mechanism through series connection of DenseASPP structure and channel attention mechanism is added into the DeepLabv3+ encoder, and the feature map weighting module is added into the DeepLabv3+ decoder to improve the network ability of feature extraction and the segmentation accuracy of the target region edge. In terms of the loss function, the improved loss obtained by combining the Huber loss and the noise-robust Dice loss is used to solve the proportional imbalance problem of multi-category pixel number, while improving the robustness against noise. The ablation experimental results show that the mean intersection over union (MIoU) of segmentation results with improved DeepLabv3+ network and with improved DeepLabv3+ network trained by the improved loss function increase by 0.46 percent and 0.99 percent respectively compared with the original DeepLabv3+ network. Proposed method effectively improves the segmentation accuracy of the target region in the monitoring ultrasonic image for HIFU therapy and has strong robustness against noise.

Lane Line Extraction in Raining Weather Images by Ridge Edge Detection with Improved MSR and Hessian Matrix

To accurately detect lane lines in road traffic images at raining weather, a edge detection based method is studied, which mainly includes four algorithms. (1) Firstly an image is enhanced by an improved Retinex algorithm; (2) Then, an algorithm based on the Hessian matrix is applied to strengthen lane lines; (3) To extract the feature points of a lane line, a ridge edge detection algorithm based on five line detection in four directions is proposed, in which, in light on the possible positions of lane lines in the image, it detects the maximum gray level points in the local area of the detecting point within the pre-set valid detection region; and (4) After the noise removal based on the minimum circumscribed rectangles, the candidate points of lane lines are connected as segments, and for the gap filling between segments, in order to make connection correctly, the algorithm makes the filling in two steps, short gap and long gap fillings, and the long gap filling is made on the combination of segment angle difference and gap distance and gap angle. By testing hundreds of images of the lane lines at raining weather and by comparing several traditional image enhancement and segmentation algorithms, the new method of the lane line detection can produce the satisfactory results.