At present, computed tomography (CT) is widely used to assist disease diagnosis. Especially, computer aided diagnosis (CAD) based on artificial intelligence (AI) recently exhibits its importance in ...intelligent healthcare. However, it is a great challenge to establish an adequate labeled dataset for CT analysis assistance, due to the privacy and security issues. Therefore, this paper proposes a convolutional autoencoder deep learning framework to support unsupervised image features learning for lung nodule through unlabeled data, which only needs a small amount of labeled data for efficient feature learning. Through comprehensive experiments, it shows that the proposed scheme is superior to other approaches, which effectively solves the intrinsic labor-intensive problem during artificial image labeling. Moreover, it verifies that the proposed convolutional autoencoder approach can be extended for similarity measurement of lung nodules images. Especially, the features extracted through unsupervised learning are also applicable in other related scenarios.
Violent interaction detection is a hot topic in computer vision. However, the recent research works on violent interaction detection mainly focus on the traditional hand-craft features, and does not ...make full use of the research results of deep learning in computer vision. In this paper, we propose a new robust violent interaction detection framework based on multi-stream deep learning in surveillance scene. The proposed approach enhances the recognition performance of violent action in video by fusing three different streams: attention-based spatial RGB stream, temporal stream, and local spatial stream. The attention-based spatial RGB stream learns the spatial attention regions of persons that have high probability to be action region through soft-attention mechanism. The temporal stream employs optical flow as input to extract temporal features. The local spatial stream learns spatial local features using block images as input. Experimental results demonstrate the effectiveness and reliability of the proposed method on three violent interactive datasets: hockey fights, movies, violent interaction. We also verify the proposed method on our own elevator surveillance video dataset and the performance of the proposed method is satisfied.
Recent salient object detection methods are mainly based on Convolutional Neural Networks (CNNs). Most of them adopt a U-shape architecture to extract and fuse multi-scale features. The coarser-level ...semantic information is progressively transmitted to finer-level layers through continuous upsampling operations, and the coarsest-level features will be diluted, resulting in the salient object boundary being blurred. On the other hand, the hand-craft feature has the advantage of being purposive and easy to calculate, in which the edge density feature may help improve the sharpness of the salient object boundary by the rich edge information. In this paper, we propose a Co-Guided Attention Network (CoGANet). On the base of the Feature Pyramid Network (FPN), our model implements a co-guided attention mechanism between the image itself and its edge density feature. In the bottom-up pathway of FPN, two streams separately work, taking the original image and its edge density feature as inputs, and each producing five feature maps. Then the last feature map in each stream generates a set of attention maps through a Multi-scale Spatial Attention Module (MSAM). In the top-down pathway, the attention maps of one stream are directly delivered to each stage in the other stream. These attention maps are fused with the feature maps by an Attention-based Feature Fusion Module (AFFM). Finally, an accurate saliency map is produced by fusing the finest-level outputs of the two streams. Experimental results on five benchmark datasets demonstrate our model is superior to 13 state-of-the-art methods in terms of four evaluation metrics.
This paper proposes BoF-CP approach which is a novel scheme feature fusion for writer-dependent offline signature verification. At the heart of the new methodology lies feature extraction in ...surrounded candidate points. In the proposed method, at first, several features of the type of component-oriented and pixel-oriented features are extracted at the region of candidate points. Due to the different geometric structure in different signatures, several short feature vectors are created to the number of candidate points in each image. In the proposed approach, the corresponding homogeneous feature vectors are fused based on standard deviation and variance at the candidate points to create a normalized vector. We called the proposed method Bag of Features in Candidate Point (BoF-CP). Finally, the normalized feature vector enters the classification stage to verify the query samples. To evaluate the proposed method, standard datasets MCYT, GPDS, and CEDAR have been used. According to the experimental results, the proposed approach has been able to import optimal features into the classification algorithm by which the recognition rate in the separation of the genuine and forgery samples has been enhanced. According to the obtained statistical results, the proposed method has improved in several criteria such as average error rate, accuracy, sensitivity, specify as compared to state of the art.
Dalbergia tonkinensis (Fabaceae) is one of the most valuable rosewood furniture species in the world and, although threatened in the wild, is being traded in local markets in north Vietnam. Trade in ...D. tonkinensis was investigated by compiling information from interviews, published and unpublished articles, and reports of forest rangers and market regulators. Goods traded included seed, seedlings, living trees, heartwood, handcrafts and furniture. Trees for transplanting ranged in value from 43.5-87.0 USD for a small tree to 434.8-4,347.8 USD for larger trees. The price of heartwood varied greatly depending on size, from 13.0 USD/kg (diameter = 6 cm) to 152.2 USD/kg (diameter = 20 cm). High quality furniture sold for 0.4-1.1 million USD. In the period 2011 to 2018, 905.2 tons of heartwood was traded valued at 16.5 million USD. Two thirds of D. tonkinensis wood being traded came from home gardens in 18 provinces. Vinh Phuc province was the largest provider of trees and also had networks of wood purchasing and processing villages. There is now a more open policy towards an emerging D. tonkinensis industry in Vietnam and regulations encourage communities and households to plant D. tonkinensis.
Vision-based traffic accident detection is one of the challenging tasks in intelligent transportation systems due to the multi-modalities of traffic accidents. The first challenging issue is about ...how to learn robust and discriminative spatio-temporal feature representations. Since few training samples of traffic accidents can be collected, sparse coding techniques can be used for small data case. However, most sparse coding algorithms which use norm regularisation may not achieve enough sparsity. The second challenging issue is about the sample imbalance between traffic accidents and normal traffic such that detector would like to favour normal traffic. This study proposes a traffic accident detection method, including a self-tuning iterative hard thresholding (ST-IHT) algorithm for learning sparse spatio-temporal features and a weighted extreme learning machine (W-ELM) for detection. The ST-IHT algorithm can improve the sparsity of encoded features by solving an norm regularisation. The W-ELM can put more focus on traffic accident samples. Meanwhile, a two-point search strategy is proposed to adaptively find a candidate value of Lipschitz coefficients to improve the tuning precision. Experimental results in our collected dataset have shown that this proposed traffic accident detection algorithm outperforms other state-of-the-art methods in terms of the feature's sparsity and detection performance.
Kotaka Otsuma is an author of many books whose topics range from school education to housewifery and women in general. She has also broadened the basis of home economics education in the Taisho to ...Showa period. Currently, among her works, 31 books from the library of Otsuma Women's University Library have been digitally archived and published as the “Kotaka Otsuma’s Collection”. Although this collection is important to understand the achievements of Kotaka Otsuma and various aspects of school education in her time, it is not well known to the public and has not been utilized effectively due to the website that was not designed based on an efficient browsing method. This study aimed to propose a method to use the “Kotaka Otsuma’s Collection” effectively through the design of a new website and the restoration of the works by students. As for the method, I first examined five existing websites that use digital archives and created three types of test sites: a museum database type, a social media type, and a general homepage type. Based on a survey on the three sites for students, I found that the general homepage type with many images is suitable for students to perceive the general outline of “Kotaka Otsuma’s Collection” because they can view and browse the content easily. Also, linking multiple websites increased the site recognition level, which led to an increase in the number of viewers. I also concluded that it is useful to actively incorporate digital archives in educational settings, such as work restoration, as a means to deepen the students’ learning experiences.
In order to detect early stomach malignancies, upper GI endoscopy is commonly used. An object detection model, a form of deep learning, was projected as a means of automating the diagnosis of early ...stomach cancer using endoscopic pictures. Yet there were difficulties in reducing false positives in the results that were detected. Thus, this research proposes an automatic classification method for stomach cancer classification using a (CNN) that has been pre-trained. To classify cancer in endoscopic pictures automatically, our method is superior to those that rely on traditional, manual features. Thirteen convolutional layers with small 33 size kernels and three fully linked layers make up the suggested model. We use the learning approach, which involves pre-training the weight of fine-tuning the weight of layers with the Slime Mould Algorithm, to deal with the data scarcity (SMA). Experiments employing 1208 photos from healthy subjects and 533 photographs from patients with stomach cancer examined detection performance using the 5-fold cross validation approach. These findings suggest the projected strategy will be effective for automating the diagnosis of early stomach cancer in endoscopic pictures.
PDF
