Shape-Consistent One-Shot Unsupervised Domain Adaptation for Rail Surface Defect Segmentation

The techniques of eigenfaces and neural netbased algorithms (LS-SVM and BP NNs) are combined to categorize gender from facial images in this paper. Based on exploration of the ralated techniques, the eigenfaces were firstly established from the training images, and the projection coefficients for training and testing images obtained in the space spanned by the eigenfaces; after that the LS-SVM and BP classifiers are built with training coefficients, which are used for classifying training and testing images, and classification accuracy percentage values are calculated. The experiments are implemented with our self-made facial images, and the results demonstrate that LS-SVM classification has better performance than BP . In experiments we also use cross validation to determine the number of selected primary components and kernel function parameter.

2D cine MR imaging may be utilized to monitor rapidly moving tumors and organs-at-risk for real-time adaptive radiotherapy. This study systematically investigates the impact of geometric imaging parameters on the ability of 2D cine MR imaging to guide template-matching-driven autocontouring of lung tumors and abdominal organs.Abdominal 4D MR images were acquired of six healthy volunteers and thoracic 4D MR images were obtained of eight lung cancer patients. At each breathing phase of the images, the left kidney and gallbladder or lung tumor, respectively, were outlined as volumes of interest. These images and contours were used to create artificial 2D cine MR images, while simultaneously serving as 3D ground truth. We explored the impact of five different imaging parameters (pixel size, slice thickness, imaging plane orientation, number and relative alignment of images as well as strategies to create training images). For each possible combination of imaging parameters, we generated artificial 2D cine MR images as training and test images. A template-matching algorithm used the training images to determine the tumor or organ position in the test images. Subsequently, a 3D base contour was shifted to the determined position and compared to the ground truth via centroid distance and Dice similarity coefficient.The median centroid distance between adapted and ground truth contours was 1.56 mm for the kidney, 3.81 mm for the gallbladder and 1.03 mm for the lung tumor (median Dice similarity coefficient: 0.95, 0.72 and 0.93). We observed that a decrease in image resolution led to a modest decrease in localization accuracy, especially for the small gallbladder. However, for all volumes of interest localization accuracy varied substantially more between subjects than due to the different imaging parameters.Automated tumor and organ localization using 2D cine MR imaging and template-matching-based autocontouring is robust against variation of geometric imaging parameters. Future work and optimization efforts of 2D cine MR imaging for real-time adaptive radiotherapy is needed to characterize the influence of sequence- and anatomical site-specific imaging contrast.

Sparse representation with multi-manifold analysis for texture classification from few training images

Deep neural network is a technique of deep learning, where deep neural network model have multiple hidden layers with input and output layer, but artificial neural network have single hidden layer between input and output layer. The use of multiple hidden layers in deep neural network is to improve the performance of model and achieving the higher accuracy compare to machine learning models and their accuracy. The field of pattern recognition is mostly used by the researchers for their research work. There are lots of pattern are available in the field of pattern recognition like: handwritten digits, characters, images, faces, sound, speech etc. In this paper we have concentrated on handwritten digits classification and recognition. For handwritten digit datasets, we have used commonly known Arkiv Digital Sweden (ARDIS) [1] dataset and United State postal service (USPS) [7] dataset. ARDIS dataset is a collection of 7600 samples, where 6600 used as training samples and 1000 used as testing samples. USPS dataset is a collection of 10000 image samples where 7291 samples are used as training sample and 2007 samples are used as testing samples. In this paper we have implemented the proposed deep neural network technique for the classification and recognition of the ARDIS and USPS dataset. The proposed model has collection of 6 layers with relu and softmax activation function. After implementing model, 98.70% testing and 99.76% training accuracy for ARDIS samples achieved, which is higher than previous research accuracy. Also 98.22% training and 93.01%testing accuracy with USPS samples dataset has been achieved. The results represents the performance of deep neural networks have been outstanding compare to other previous techniques.

Robust segmentation of arterial walls in intravascular ultrasound images using Dual Path U-Net

This study tries to find the performance of ResNet 50, 101 and 152 version 1 architecture to classify normal and pneumonia using chest X-ray. For the second goal, the effect of image dimension is investigated. Furthermore, the different distribution of training, validation and testing images in the dataset is explored, i.e. dataset A and B. 5,838 chest X-ray images are used in dataset A. For dataset B, 5,856 images are used. These images are distributed randomly as training, validation and testing images with compositions of 80%, 10% and 10%. For each dataset, 12 image dimensions are used, i.e., 96 × 96; 128 × 128; 160 × 160; 192 × 192; 224 × 224; 256 × 256; 288 × 288; 320 × 320; 352 × 352; 384 × 384; 416 × 416 and 448 × 448 pixels. Moreover, this study used exposure fusion framework as image enhancement in the preprocessing stage. In general, ResNet 152 architecture has better performance than ResNet 50 and 152. However, the best performance is achieved by dataset A using ResNet 50 and an image dimension of 384 × 384 pixels. This model has an accuracy, sensitivity, specificity, precision, F1-score and ROC of 97.6%, 99.3%, 93.1%, 97.5%, 98.4% and 96.2%. The pneumonia detection using ResNet is affected by the distribution of training, validation and testing images. Furthermore, there is no correlation between the image dimension and the detection performance.

Abstract. We propose an image labeling method for LIDAR intensity image obtained by Mobile Mapping System (MMS) using K-Nearest Neighbor (KNN) of feature obtained by Convolutional Neural Network (CNN). Image labeling assigns labels (e.g., road, cross-walk and road shoulder) to semantic regions in an image. Since CNN is effective for various image recognition tasks, we try to use the feature of CNN (Caffenet) pre-trained by ImageNet. We use 4,096-dimensional feature at fc7 layer in the Caffenet as the descriptor of a region because the feature at fc7 layer has effective information for object classification. We extract the feature by the Caffenet from regions cropped from images. Since the similarity between features reflects the similarity of contents of regions, we can select top K similar regions cropped from training samples with a test region. Since regions in training images have manually-annotated ground truth labels, we vote the labels attached to top K similar regions to the test region. The class label with the maximum vote is assigned to each pixel in the test image. In experiments, we use 36 LIDAR intensity images with ground truth labels. We divide 36 images into training (28 images) and test sets (8 images). We use class average accuracy and pixel-wise accuracy as evaluation measures. Our method was able to assign the same label as human beings in 97.8% of the pixels in test LIDAR intensity images.

As a key factor, the availability of large-scale training samples determines the improvement of visual performance. However, the size of Fine-Grained Visual Categorization (FGVC) datasets is always limited. Therefore, overfitting as an issue in FGVC-related training needs to be solved. Data mixing augmentation is a widely-used data augmentation method. In most of the recently proposed data mixing augmentation methods, random patch selection may generate meaningless training samples and result in model instability during the training process. This paper proposes a data mixing augmentation strategy termed Semantic and Attentive Data Mixing (SADMix) to select semantic patches for the generation of new training samples. In SADMix, a certain number of critical regions are localized according to convolutional activations. An image patch is selected from these localized regions for the generation of new training samples. The size, aspect ratio, and center location of these image patches are changed according to the random values from a beta distribution. These image patches with semantic information are used to mix two training images. According to the class activation map (CAM), training images and their labels are mixed proportionally to generate new mixed training samples and the corresponding labels. The proposed SADMix is tested on three fine-grained datasets, which are CUB-200-2011, FGVC Aircraft, and Stanford Cars, respectively. The experimental results confirm the effectiveness of the proposed SADMix.

Noninvasive assessment of diastolic dysfunction relies on multiple echocardiographic indicators, including measurements from both standard B-mode images and Doppler, obtained at various cardiac locations such as the mitral annulus, tricuspid annulus, left ventricle, and left atrium. The diagnostic process is complex and subject to interobserver variability, making accurate and rapid evaluation challenging. Automated semantic segmentation of key cardiac structures, such as the left atrium, left ventricle, and mitral valve annulus, offers a potential solution by capturing temporal changes throughout the cardiac cycle. This study aims to improve the accuracy of segmenting the left atrium, left ventricle, and mitral valve annulus in echocardiographic images and to leverage the resulting temporal segmentation features for more reliable identification of diastolic dysfunction. This study presents Diff-TransUNet, a novel segmentation model incorporating a noise-robust Differential Transformer module. Evaluations on private (1137 training images, 135 validation images, and 88 test images), CAMUS (1400 training images, 200 validation images, and 200 test images), and EchoNet-Dynamic (5000 training images, 2546 validation images, and 2528 test images) datasets demonstrate improved performance over state-of-the-art methods, assessed by Dice coefficient (Dice), Intersection-over-Union (IoU), and 95th percentile Hausdorff Distance (HD95) metrics. Statistical analysis was performed to compare Diff-TransUNet with baseline methods across evaluation metrics. To control for errors arising from multiple comparisons, p-values were adjusted using the Benjamini-Hochberg false discovery rate (FDR) correction. Statistical significance was assessed at a 95% confidence level. In addition to p-values, Cohen's d effect size was computed to quantify the practical significance of performance differences. The proposed Diff-TransUNet achieved a Dice of 87.49%, IoU of 79.07%, and HD95 of 1.48 on the private dataset. Compared with state-of-the-art models, Dice improved by 1.35%-4.30% (p<0.05, Cohen's d=0.32-0.90), IoU by 1.97%-5.67% (p<0.05, Cohen's d=0.37-1.03), and HD95 by 0.16-0.83 (p<0.05, Cohen's d=0.21-0.90). On the CAMUS dataset, the model achieved a Dice of 88.74%, IoU of 80.58%, and HD95 of 2.83, showing improvements of 1.07%-4.96% (p<0.05, Cohen's d=0.18-0.63) in Dice, 1.55%-6.89% (p<0.05, Cohen's d=0.19-0.71) in IoU, and 0.41-2.85 (p<0.05, Cohen's d=0.12-0.46) in HD95 compared to advanced models. On the EchoNet-Dynamic dataset, the model obtained a Dice of 92.25%, IoU of 85.87%, and HD95 of 1.65, outperforming other methods by 0.42%-2.00% (p<0.05, Cohen's d=0.10-0.40) in Dice, 0.69%-3.21% (p<0.05, Cohen's d=0.10-0.43) in IoU, and 0.21-1.12 (p<0.05, Cohen's d=0.09-0.34) in HD95. Furthermore, by extracting volumetric segmentation features, the proposed method achieved an accuracy of 88.95% (95 % CI 87.15% to 90.08%) in identifying diastolic dysfunction. The proposed Diff-TransUNet model achieves significant improvements in ultrasound segmentation. Features extracted from the left ventricle, left atrium, and mitral annulus segmented by Diff-TransUNet can be effectively used for the identification of diastolic dysfunction.

Background: Legumes play an important role in improving soil quality and nutrition in humans. Soybean is one of the legumes which are rich in protein and oil content. Nutrients deficiency in soybean plant could affect the growth of the plants and might lead to loss in its yield. The developments in modern technologies such as computer vision and deep learning are being leveraged in identifying nutrient deficiencies in soybean plants. Methods: Convolutional neural network with six feature extraction blocks and one classification block is developed to identify macro nutrient deficiency in soybean plants. Images are first collected, pre-processed and labeled. They are then split into training images and testing images in the ratio of 80:20. The proposed convolutional neural network model is trained over training images and the final trained model is tested with testing images. Result: In detecting the deficiency of macro nutrients such as nitrogen, phosphorus and potassium in soybean plants, the testing results of the proposed convolutional neural network architecture achieved an accuracy of 97.43%. Accuracy comparison against existing models such as VGG16, ResNet50 and MobileNetV3 demonstrate that the proposed model effectively identifies nutrient deficiencies in soybean plants. Thus, the proposed system is designed to support farmers in making timely decisions and to contribute to food security by leveraging deep learning techniques.

We propose a novel method for weakly supervised semantic segmentation. Training images are labeled only by the classes they contain, not by their location in the image. On test images instead, the method predicts a class label for every pixel. Our main innovation is a multi-image model (MIM) - a graphical model for recovering the pixel labels of the training images. The model connects superpixels from all training images in a data-driven fashion, based on their appearance similarity. For generalizing to new test images we integrate them into MIM using a learned multiple kernel metric, instead of learning conventional classifiers on the recovered pixel labels. We also introduce an “objectness” potential, that helps separating objects (e.g. car, dog, human) from background classes (e.g. grass, sky, road). In experiments on the MSRC 21 dataset and the LabelMe subset of [18], our technique outperforms previous weakly supervised methods and achieves accuracy comparable with fully supervised methods.

For the submission from the University of Glasgow for the Image-CLEF 2009 Robot Vision Task a large set of interesting points were extracted using an edge corner detector, these points were used to represent each image. The RANSAC method [1] was then applied to estimate the similarity between test and training images based on the number of matched pairs of points. The location of robot was then annotated based on the training image which contains the highest number of matched point pairs with the test image. A set of decision rules with the respect to the trajectory behaviour of robot's motion were defined to refine the final results. An illumination filter was also applied for two of the runs in order to reduce the illumination effect.

Same-different visual reasoning is a basic skill central to abstract combinatorial thought. This fact has lead neural networks researchers to test same-different classification on deep convolutional neural networks (DCNNs), which has resulted in a controversy regarding whether this skill is within the capacity of these models. However, most tests of same-different classification rely on testing on images that come from the same pixel-level distribution as the training images, yielding the results inconclusive. In this study, we tested relational same-different reasoning in DCNNs. In a series of simulations we show that models based on the ResNet architecture are capable of visual same-different classification, but only when the test images are similar to the training images at the pixel level. In contrast, when there is a shift in the testing distribution that does not change the relation between the objects in the image, the performance of DCNNs decreases substantially. This finding is true even when the DCNNs’ training regime is expanded to include images taken from a wide range of different pixel-level distributions or when the model is trained on the testing distribution but on a different task in a multitask learning context. Furthermore, we show that the relation network, a deep learning architecture specifically designed to tackle visual relational reasoning problems, suffers the same kind of limitations. Overall, the results of this study suggest that learning same-different relations is beyond the scope of current DCNNs.

A large number of automatic face recognition systems have been reported in the literature. Many of them are robust to within class appearance variation of subjects such as variation in expression, lighting and pose. However, most of the face identification systems developed, are sensitive to changes in the age of individuals. In this paper we demonstrate that automatic age simulation techniques can be used for designing face recognition systems, robust to ageing variation. In this context, the perceived age of the subjects in the training and test images is modified before the training and classification procedures, so that ageing variation is eliminated. Experimental results demonstrate that the performance of our face recognition system can be improved significantly, when this approach is adopted.

The process of aging causes significant alterations in the facial appearance of individuals. When compared with other sources of variation in face images, appearance variation due to aging displays some unique characteristics. Changes in facial appearance due to aging can even affect discriminatory facial features, resulting in deterioration of the ability of humans and machines to identify aged individuals. We describe how the effects of aging on facial appearance can be explained using learned age transformations and present experimental results to show that reasonably accurate estimates of age can be made for unseen images. We also show that we can improve our results by taking into account the fact that different individuals age in different ways and by considering the effect of lifestyle. Our proposed framework can be used for simulating aging effects on new face images in order to predict how an individual might look like in the future or how he/she used to look in the past. The methodology presented has also been used for designing a face recognition system, robust to aging variation. In this context, the perceived age of the subjects in the training and test images is normalized before the training and classification procedure so that aging variation is eliminated. Experimental results demonstrate that, when age normalization is used, the performance of our face recognition system can be improved.