Search

Epilepsy is one kind of neurological disease characterized by recurring seizures. Recurrent seizures can cause ongoing negative mental and cognitive damage to the patient. Therefore, timely diagnosis and treatment of epilepsy are crucial for patients. Manual electroencephalography (EEG) signals analysis is time and energy consuming, making automatic detection using EEG signals particularly important. Many deep learning algorithms have thus been proposed to detect seizures. These methods rely on expensive and bulky hardware, which makes them unsuitable for deployment on devices with limited resources due to their high demands on computer resources. In this paper, we propose a novel lightweight neural network for seizure detection using pure convolutions, which is composed of inverted residual structure and multi-scale channel attention mechanism. Compared with other methods, our approach significantly reduces the computational complexity, making it possible to deploy on low-cost portable devices for seizures detection. We conduct experiments on the CHB-MIT dataset and achieves 98.7% accuracy, 98.3% sensitivity and 99.1% specificity with 2.68 M multiply-accumulate operations (MACs) and only 88 K parameters. [ABSTRACT FROM AUTHOR]

The proliferation of deep learning has driven research into deep learning‐based logo detection, which usually needs a large number of annotated data to train the model. However, due to the occasional appearance of new brands or the high cost of annotation, the number of training data is limited. Against this backdrop, the authors adapt the few‐shot object detection into logo detection, and thus present a cutting‐edge method called Double Classification Head (DCH) for Few‐Shot Logo Detection (DCH‐FSLogo), which aims at detecting the unseen logo classes using few annotated data. Unlike the traditional few‐shot detection, some logo objects are similar to their backgrounds and have diverse shapes as well. For this reason, the authors adopt balanced feature pyramid and deformable Region of Interest pooling in DCH‐FSLogo, this enhances the feature extraction capability and adapts to the different logo shapes. In addition, we introduce the DCH for few‐shot logo detection to detect logo objects using few annotated data. Specifically, we use an extra classification head for the base classes to ease the influence from the novel classes. The experimental results on four datasets, namely: FlickrLogos‐32, FoodLogoDet‐1500‐100, LogoDet‐3K‐100 and QMUL‐OpenLogo‐100, demonstrate that our method achieves better performance. [ABSTRACT FROM AUTHOR]

Logo detection is a technology that identifies logos in images and returns their locations. With logo detection technology, brands can check how often their logos are displayed on social media platforms and elsewhere online and how they appear. It has received a lot of attention for its wide applications across different sectors, such as brand identity protection, product brand management, and logo duration monitoring. Particularly, logo detection technology can offer various benefits for companies to help brands measure their logo coverage, track their brand perception, secure their brand value, increase the effectiveness of their marketing campaigns and build brand awareness more effectively. However, compared with the general object detection, logo detection is more challenging due to the existence of both small logo objects and large aspect ratio logo objects. In this paper, we propose a novel approach, named Discriminative Semantic Feature Pyramid Network with Guided Anchoring (DSFP-GA), which can address these challenges via aggregating the semantic information and generating different aspect ratio anchor boxes. More specifically, our approach mainly consists of two components, namely Discriminative Semantic Feature Pyramid (DSFP) and Guided Anchoring (GA). The former is proposed to fuse semantic features into low-level feature maps to obtain discriminative representation of small logo objects, while the latter is further integrated into DSFP to generate large aspect ratio anchor boxes for detecting large aspect ratio logo objects. Extensive experimental results on four benchmarks demonstrate the effectiveness of the proposed DSFP-GA. Moreover, we further conduct visual analysis and ablation studies to illustrate the strength of the proposed DSFP-GA when detecting both small logo objects and large aspect logo objects. [ABSTRACT FROM AUTHOR]

Logo detection is one of the crucial branches in computer vision due to various real-world applications, such as automatic logo detection and recognition, intelligent transportation, and trademark infringement detection. Compared with traditional handcrafted-feature-based methods, deep learning-based convolutional neural networks (CNNs) can learn both low-level and high-level image features. Recent decades have witnessed the great feature representation capabilities of deep CNNs and their variants, which have been very good at discovering intricate structures in high-dimensional data and are thereby applicable to many domains including logo detection. However, logo detection remains challenging, as existing detection methods cannot solve well the problems of a multiscale and large aspect ratios. In this paper, we tackle these challenges by developing a novel long-range dependence involutional network (LDI-Net). Specifically, we designed a strategy that combines a new operator and a self-attention mechanism via rethinking the intrinsic principle of convolution called long-range dependence involution (LD involution) to alleviate the detection difficulties caused by large aspect ratios. We also introduce a multilevel representation neural architecture search (MRNAS) to detect multiscale logo objects by constructing a novel multipath topology. In addition, we implemented an adaptive RoI pooling module (ARM) to improve detection efficiency by addressing the problem of logo deformation. Comprehensive experiments on four benchmark logo datasets demonstrate the effectiveness and efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]

Logo detection involves the tasks of locating and classifying logo objects in images and videos, and has been widely applied in the real world. However, most existing approaches rely on general object detection strategies that do not fully utilize the unique characteristics of logos. This can lead to sub-optimal performance in complex environments, especially when logos are small or have varying sizes and shapes. We observe that logos belonging to the same category often share similar context information, such as background dependency of the logo. This motivates us to incorporate contextual information to improve logo detection. Our proposed method, Context-based Modeling Enhancement Network (CME-Net), aims to enhance the distinctive region feature of logos using contextual information. We achieve this by modeling both the logo and its background region to extract their contextual information. This contextual information serves as a guide for enhancing the saliency of the distinctive regions within the logo image. To further improve the accuracy of detection, we have implemented a scale feature balance strategy. This strategy solves the problem of losing scale information caused by enhancement, ensuring that all scales are appropriately considered. Additionally, noise generated during the enhancement process is also effectively suppressed. By effectively leveraging contextual information, our method successfully tackles the challenge of accurately locating logo objects. Our extensive experiments on four public benchmark datasets demonstrate that CME-Net improves the accuracy of logo detection in complex environments. • A Context-based Modeling Enhancement Network (CME-Net) is proposed to address the problem of difficult logo detection in complex environments. • A targeted feature enhancement process guided by contextual information is implemented. • CME-Net achieved great performance on four logo detection datasets. [ABSTRACT FROM AUTHOR]

The recent proliferation of advertising (ad) videos has driven the research in multiple applications, ranging from video analysis to video indexing and retrieval. Among them, classifying ad video is a key task because it allows automatic organization of videos according to categories or genres, and this further enables ad video indexing and retrieval. However, classifying ad video is challenging compared to other types of video classification because of its unconstrained content. While many studies focus on embedding ads relevant to videos, to our knowledge, few focus on ad video classification. In order to classify ad video, this paper proposes a novel ad video representation that aims to sufficiently capture the latent semantics of video content from multiple views in an unsupervised manner. In particular, we represent ad videos from four views, including bag-of-feature (BOF), vector of locally aggregated descriptors (VLAD), fisher vector (FV) and object bank (OB). We then devise a multi-layer multi-view topic model, mlmv_LDA , which models the topics of videos from different views. A topical representation for video, supporting category-related task, is finally achieved by the proposed method. Our empirical classification results on 10,111 real-world ad videos demonstrate that the proposed approach effectively differentiate ad videos. [ABSTRACT FROM AUTHOR]

The recent proliferation of videos has brought out the need for applications such as automatic annotation and organization . These applications could greatly benefit from the respective thematic content depending on the type of video. Unlike the other kinds of video, an advertising video usually conveys a specific theme in a certain time period (e.g. drawing the audience׳s attention to a product or emphasizing the brand). Traditional multi-label algorithms may not work effectively with advertising videos due mainly to their heterogeneous nature. In this paper, we propose a new learning paradigm to resolve the problems arising out of traditional multi-label learning in advertising videos through label relevance. Aiming to address the issue of label relevance, we firstly assign each label with label degree ( LD ) to classify all the labels into three groups such as first label ( FL ), important label ( IL ) and common label ( CL ), and then propose a Directed Probability Label Graph ( DPLG ) model to mine the most related labels from the multi-label data with label relevance, in which the interdependency between labels is considered. In the implementation of DPLG , the labels that appear occasionally and possess inconspicuous co-occurrences are consequently eliminated effectively, employing λ - filtering and τ - pruning processes, respectively. And then the graph theory is utilized in DPLG to acquire Correlative Label-Sets ( CLSs ). Lastly, the searched Correlative Label-Sets ( CLSs ) are utilized to enhance multi-label annotation. Experimental results on advertising videos and several publicly available datasets demonstrate the effectiveness of the proposed method for multi-label annotation with label relevance. [ABSTRACT FROM AUTHOR]

• We design a new strong baseline method based on an adaptive representation algorithm, called obtaining sufficient features for logo (OSF-Logo), which aims to solve the logo deformation and multi-scale problem. • Our method adds a modulated deformable convolution module with adaptive offsets and amplitudes to more convolution layers in the stage of feature extraction. • A set of efficient feature sampling operations is introduced into FPN. Through feature reorganization of prediction kernel, the features of each layer can be fully integrated, which can adaptively obtain more comprehensive feature content, and effectively solve the deformation and multi-scale problem of logo images. • We conduct extensive experiment and evaluations to verify the effectiveness of our method on four different scale benchmark datasets. Our method has achieved good results. The proliferation of logo has driven research into multiple applications, like logo duration monitoring in advertising videos and logo infringement detection. Recently, logo research has attracted a rapt attention and keen interest from researchers. In these studies, logo detection is challenging due to the characteristics of the logo (like multi-scale, large-scale categories, viewpoint and part deformation etc.) and the complexity of its background. In this paper, we design a new strong baseline method based on an adaptive representation algorithm, called obtaining sufficient features for logo (OSF-Logo). The method aims to address the challenges of the multi-scale objects, large-scale categories, viewpoint logos and part deformation via introducing two modules. Specifically, we introduce a regulated deformable convolution module with offsets and amplitudes to more convolution layers in the stage of feature extraction. In addition, we add an up-sampling operator to FPN for aggregating information into a large receptive field. The experimental results on several publicly available datasets demonstrate the effectiveness of OSF-Logo. [ABSTRACT FROM AUTHOR]

• The attention modules are constructed to extract and fuse RGB-D data effectively. • A fusion branch and the bi-direction multi-step propagation strategy are introduced. • Extensive experiments are performed on public NYUDv2 and SUN-RGBD datasets. Semantic segmentation based on the complementary information from RGB and depth images has recently gained great popularity, but due to the difference between RGB and depth maps, how to effectively use RGB-D information is still a problem. In this paper, we propose a novel RGB-D semantic segmentation network named RAFNet, which can selectively gather features from the RGB and depth information. Specifically, we construct an architecture with three parallel branches and propose several complementary attention modules. This structure enables a fusion branch and we add the Bi-directional Multi-step Propagation (BMP) strategy to it, which can not only retain the feature streams of the original RGB and depth branches but also fully utilize the feature flow of the fusion branch. There are three kinds of complementary attention modules that we have constructed. The RGB-D fusion module can effectively extract important features from the RGB and depth branch streams. The refinement module can reduce the loss of semantic information and the context aggregation module can help propagate and integrate information better. We train and evaluate our model on NYUDv2 and SUN-RGBD datasets, and prove that our model achieves state-of-the-art performances. [ABSTRACT FROM AUTHOR]

• Vision-based food nutrient estimation promoted development of rapid analysis of food nutrition. • Foods nutrition assessment method based on the RGB-D fusion achieved better performance. • The RGB-D fusion network combined multimodal feature fusion and multi-scale fusion modules. • The percentage values of the mean absolute error for calories and mass were only 15.0% and 10.8%. With the development of deep learning technology, vision-based food nutrition estimation is gradually entering the public view for its advantage in accuracy and efficiency. In this paper, we designed one RGB-D fusion network, which integrated multimodal feature fusion (MMFF) and multi-scale fusion for visioin-based nutrition assessment. MMFF performed effective feature fusion by a balanced feature pyramid and convolutional block attention module. Multi-scale fusion fused different resolution features through feature pyramid network. Both enhanced feature representation to improve the performance of the model. Compared with state-of-the-art methods, the mean value of the percentage mean absolute error (PMAE) for our method reached 18.5%. The PMAE of calories and mass reached 15.0% and 10.8% via the RGB-D fusion network, improved by 3.8% and 8.1%, respectively. Furthermore, this study visualized the estimation results of four nutrients and verified the validity of the method. This research contributed to the development of automated food nutrient analysis (Code and models can be found at http://123.57.42.89/codes/RGB-DNet/nutrition.html). [ABSTRACT FROM AUTHOR]

Fruit detection and segmentation will be essential for future agronomic management, with applications in yield estimation, growth monitoring, intelligent picking, disease detection and etc. In order to more accurately and efficiently realize the recognition and segmentation of apples in natural orchards, a robust segmentation net framework specially developed for fruit production is proposed. This model was improved for the more challenging problem which segments the overlapped apples from the monochromatic background regardless of various corruptions. The method extends Mask R-CNN by embedding an attention mechanism for focusing more on the informative pixels but also suppressing the noise caused by adverse factors (occlusions, overlaps, etc.), which could be more suitable and robust for operating in complex natural environment. Specifically, the Gaussian non-local attention mechanism is transplanted into Mask R-CNN for refining the semantic features generated continuously by residual network and feature pyramid network, then the model forward processing based on the balanced feature levels and finally segments the regions where the apples are located. Experimental results verify the hypothesis of current work and show that the proposed method outperforms other start-of-the-art detection and segmentation models, the AP box and AP mask metric values have reached 85.6% and 86.2% in a reasonable run time, respectively, which can meet the precision and robustness of vision system in agronomic management. [ABSTRACT FROM AUTHOR]

The correlation filters based trackers have achieved an excellent performance for object tracking in recent years. However, most existing methods use only one filter but ignore the information of the previous filters. In this paper, we propose a novel online multi-expert learning algorithm for visual tracking. In our proposed scheme, there are former trackers which retain the previous filters, and those trackers will give their predictions in each frame. The current tracker represents the filter of current frame, and both the current tracker and the former trackers constitute our expert ensemble. We use an adaptive Second-order Quantile strategy to learn the weights of each expert, which can take full advantage of all the experts. To simplify our model and remove some bad experts, we prune our models via a minimum entropy criterion. Finally, we propose a new update strategy to avoid the model corruption problem. Extensive experimental results on both OTB2013 and OTB2015 benchmarks demonstrate that our proposed tracker performs favorably against state-of-the-art methods. [ABSTRACT FROM AUTHOR]

In the field of face recognition, a key issue is whether there are a sufficient number of face training samples with valid information. Due to the complexity of human face images, face recognition is easy to be affected by the external environment such as light intensity, gesture expression, hairstyle, and occlusion. Therefore, it is difficult to obtain enough effective samples in practical applications. In this paper, we propose a new algorithm that generates virtual images by utilizing the information of the test sample via singular value decomposition. The virtual images not only extend the training sample set but also can better adapt to the test sample. In addition, we use the weighted score fusion scheme to calculate the ultimate result, which can better take advantages of data from different sources including original images and virtual images. Experimental results on the Extended Yale_B, AR, GT, ORL, and FERET face databases prove that our algorithm can obtain satisfactory performance. [ABSTRACT FROM AUTHOR]

The Elman neural network has good dynamic properties and strong global stability, being most widely used to deal with nonlinear, dynamic, and complex data. However, as an optimization of the backpropagation (BP) neural network, the Elman model inevitably inherits some of its inherent deficiencies, influencing the recognition precision and operating efficiency. Many improvements have been proposed to resolve these problems, but it has proved difficult to balance the many relevant features such as storage space, algorithm efficiency, recognition precision, etc. Also, it is difficult to obtain a permanent solution from a temporary solution simultaneously. To address this, a genetic algorithm (GA) can be introduced into the Elman algorithm to optimize the connection weights and thresholds, which can prevent the neural network from becoming trapped in local minima and improve the training speed and success rate. The structure of the hidden layer can also be optimized using the GA, which can solve the difficult problem of determining the number of neurons. Most previous studies on such evolutionary Elman algorithms optimized the connection weights or network structure individually, which represents a slight deficiency. We propose herein a novel optimized GA-Elman neural network algorithm where the connection weights are real-encoded, while the neurons of the hidden layer also adopt real-coding but with the addition of binary control genes. In this new algorithm, the connection weights and the number of hidden neurons are optimized using hybrid encoding and evolution simultaneously, greatly improving the performance of the resulting novel GA-Elman algorithm. The results of three experiments show that this new GA-Elman model is superior to the traditional model in terms of all calculated indexes. [ABSTRACT FROM AUTHOR]

• A concise and accurate FoveaMask model is proposed for segmentation of green fruits. • A position attention module(PAM) is added into mask branch for improving robustness. • FoveaMask eliminates dependence on anchors. • It can be referred and even migrated directly to other fruit segmentation. In the process of agricultural automation production, efficient and accurate segmentation of target fruit is the basis and guarantee for numerous applications including crop growth monitoring, yield prognosis, and machine picking. Green fruit is apt to affect by complicated scenes such as occlusions and overlaps, as well as the homo-chromatic background, which brings great challenges to the recognition and segmentation. In this paper, we propose a novel methodology named FoveaMask for improving the robustness and generalization of green fruit segmentation framework. Features of input images are firstly extracted by ResNet and fused by FPN. The classification and bounding-box regression of each spatial position on the feature maps are then carried out directly by the way of full convolution. RoI Align layer is applied to fix feature region of proposals to the same size yet preserve exact spatial locations. Finally, instance-level fruit segmentation is realized by pixel-level classification on each proposal using embedded mask branches. The whole network architecture of the new model does not involve the related design and operation of anchor, which greatly improves the generalization ability for different shape fruits, alleviates the computing and storage resources, and balances the contradiction between accuracy and efficiency simultaneously. Additionally, a position attention module(PAM) is introduced into the embedding mask branch to aggregate the effective information pixels, which can improve the robustness of the segmentation model in the actual complex environment of the orchard. We test the model on green apple and immature persimmon data sets and the experimental results show that FoveaMask performs best in terms of both recognition accuracy and model complexity compared with other 11 different types of detection and segmentation models. [ABSTRACT FROM AUTHOR]

• U-Net is introduced into the green apple segmentation algorithm in complex environments. • Design an Edge structure including residual blocks and gated convolution to obtain the boundary semantic information. • Construct the ASPP structure, fusing the edge features and high-level features. Fruit segmentation is a critical step for fruit recognition, and key to the efficiency of a robotic vision system for harvesting and accurate orchard yield estimation. Due to the unstructured characteristics of orchard environment, the segmentation performance of traditional segmentation algorithms is insufficient for handling green target fruits in a complex environment. Deep learning algorithms bring a fresh perspective to target segmentation. This study proposes an ensemble U-Net segmentation model suitable for small sample datasets. Edge structures are designed by integrating residual blocks and gated convolutions to obtain the boundary semantic information of the target image; atrous convolutions are applied to resolve the contradiction between the resolution of the feature map and the receiving field, retain more multiscale context information and achieve target fruit segmentation. An atrous spatial pyramid pooling (ASPP) structure is applied to merge the edge features and the high-level features of U-Net. The experimental results show that the proposed method effectively improves the segmentation accuracy of the target fruit and the generalization ability of the model. The proposed method extends the application scope of the harvesting robot and orchard yield measurements, thereby providing a theoretical reference for other fruit and vegetable target fruit segmentation efforts. [ABSTRACT FROM AUTHOR]