Search

As a detection tool to identify metal or alloy, metallographic quantitative analysis has received increasing attention for its ability to evaluate quality control and reveal mechanical properties. The detection procedure is mainly operated manually to locate and characterize the constitution in metallographic images. The automatic detection is still a challenge even with the emergence of several excellent models. Benefiting from the development of deep learning, with regard to two different metallurgical structural steel image datasets, we propose two attention-aware deep neural networks, Modified Attention U-Net (MAUNet) and Self-adaptive Attention-aware Soft Anchor-Point Detector (SASAPD), to identify structures and evaluate their performance. Specifically, in the case of analyzing single-phase metallographic image, MAUNet investigates the difference between low-frequency and high-frequency and prevents duplication of low-resolution information in skip connection used in an U-Net like structure, and incorporates spatial-channel attention module with the decoder to enhance interpretability of features. In the case of analyzing multi-phase metallographic image, SASAPD explores and ranks the importance of anchor points, forming soft-weighted samples in subsequent loss design, and self-adaptively evaluates the contributions of attention-aware pyramid features to assist in detecting elements in different sizes. Extensive experiments on the above two datasets demonstrate the superiority and effectiveness of our two deep neural networks compared to state-of-the-art models on different metrics.

Pyramid architecture is a useful strategy to fuse multi-scale features in deep monocular depth estimation approaches. However, most pyramid networks fuse features only within the adjacent stages in a pyramid structure. To take full advantage of the pyramid structure, inspired by the success of DenseNet, this paper presents DCPNet, a densely connected pyramid network that fuses multi-scale features from multiple stages of the pyramid structure. DCPNet not only performs feature fusion between the adjacent stages, but also non-adjacent stages. To fuse these features, we design a simple and effective dense connection module (DCM). In addition, we offer a new consideration of the common upscale operation in our approach. We believe DCPNet offers a more efficient way to fuse features from multiple scales in a pyramid-like network. We perform extensive experiments using both outdoor and indoor benchmark datasets (i.e., the KITTI and the NYU Depth V2 datasets) and DCPNet achieves the state-of-the-art results.

Image semantic segmentation has been applied more and more widely in the fields of satellite remote sensing, medical treatment, intelligent transportation, and virtual reality. However, in the medical field, the study of cerebral vessel and cranial nerve segmentation based on true-color medical images is in urgent need and has good research and development prospects. We have extended the current state-of-the-art semantic-segmentation network DeepLabv3+ and used it as the basic framework. First, the feature distillation block (FDB) was introduced into the encoder structure to refine the extracted features. In addition, the atrous spatial pyramid pooling (ASPP) module was added to the decoder structure to enhance the retention of feature and boundary information. The proposed model was trained by fine tuning and optimizing the relevant parameters. Experimental results show that the encoder structure has better performance in feature refinement processing, improving target boundary segmentation precision, and retaining more feature information. Our method has a segmentation accuracy of 75.73%, which is 3% better than DeepLabv3+.

This paper proposes a novel approach to high-dynamic-range (HDR) imaging of dynamic scenes to eliminate ghosting artifacts in HDR images when in the presence of severe misalignment (large object or camera motion) in input low-dynamic-range (LDR) images. Recent non-flow-based methods suffer from ghosting artifacts in the presence of large object motion. Flow-based methods face the same issue since their optical flow algorithms yield huge alignment errors. To eliminate ghosting artifacts, we propose a simple yet effective alignment network for solving the misalignment. The proposed pyramid inter-attention module (PIAM) performs alignment of LDR features by leveraging inter-attention maps. Additionally, to boost the representation of aligned features in the merging process, we propose a dual excitation block (DEB) that recalibrates each feature both spatially and channel-wise. Exhaustive experimental results demonstrate the effectiveness of the proposed PIAM and DEB, achieving state-of-the-art performance in terms of producing ghost-free HDR images.

With the development of artificial intelligence and big data analytics, an increasing number of researchers have tried to use deep-learning technology to train neural networks and achieved great success in the field of vehicle detection. However, as a special domain of object detection, vehicle detection in aerial images still has made limited progress because of low resolution, complex backgrounds and rotating objects. In this paper, an improved feature-balanced pyramid network (FBPN) has been proposed to enhance the network&rsquo, s ability to detect small objects. By combining FBPN with modified faster region convolutional neural network (faster-RCNN), a vehicle detection framework for aerial images is proposed. The focal loss function is adopted in the proposed framework to reduce the imbalance between easy and hard samples. The experimental results based on the VEDIA, USCAS-AOD, and DOTA datasets show that the proposed framework outperforms other state-of-the-art vehicle detection algorithms for aerial images.

Multi-focus image fusion has become a very practical image processing task. It uses multiple images focused on various depth planes to create an all-in-focus image. Although extensive studies have been produced, the performance of existing methods is still limited by the inaccurate detection of the focus regions for fusion. Therefore, in this paper, we proposed a novel U-shape network which can generate an accurate decision map for the multi-focus image fusion. The Siamese encoder of our U-shape network can preserve the low-level cues with rich spatial details and high-level semantic information from the source images separately. Moreover, we introduce the ResBlocks to expand the receptive field, which can enhance the ability of our network to distinguish between focus and defocus regions. Moreover, in the bridge stage between the encoder and decoder, the spatial pyramid pooling is adopted as a global perception fusion module to capture sufficient context information for the learning of the decision map. Finally, we use a hybrid loss that combines the binary cross-entropy loss and the structural similarity loss for supervision. Extensive experiments have demonstrated that the proposed method can achieve the state-of-the-art performance.

Medical image fusion techniques can fuse medical images from different morphologies to make the medical diagnosis more reliable and accurate, which play an increasingly important role in many clinical applications. To obtain a fused image with high visual quality and clear structure details, this paper proposes a convolutional neural network (CNN) based medical image fusion algorithm. The proposed algorithm uses the trained Siamese convolutional network to fuse the pixel activity information of source images to realize the generation of weight map. Meanwhile, a contrast pyramid is implemented to decompose the source image. According to different spatial frequency bands and a weighted fusion operator, source images are integrated. The results of comparative experiments show that the proposed fusion algorithm can effectively preserve the detailed structure information of source images and achieve good human visual effects.

In synthetic aperture radar (SAR) images, ships are often arbitrary-oriented and densely arranged in complex backgrounds, posing enormous challenges for ship detection. However, most existing methods detect ships with horizontal bounding boxes, which leads to the redundancy of detected regions. Furthermore, the high Intersection-over-Union (IoU) between two horizontal bounding boxes of densely arranged ships can cause missing detection. In this paper, a multi-stage rotational region based network (MSR2N) is proposed to solve the above problems. In MSR2N, the rotated bounding boxes, which can reduce background noise and prevent missing detection caused by high IoUs, are utilized to represent ship regions. MSR2N consists of three modules: feature pyramid network (FPN), rotational region proposal network (RRPN), and multi-stage rotational detection network (MSRDN). First of all, the FPN is applied to combine high-resolution features with semantically strong features. Second, in RRPN, a rotation-angle-dependent strategy is employed to generate multi-angle anchors which can represent arbitrary-oriented ship regions more felicitously than horizontal anchors. Finally, the MSRDN with three sub-networks is proposed to regress proposals of ship regions stage by stage. Meanwhile, the incrementally increasing IoU thresholds are selected for resampling positive and negative proposals in sequential stages of MSRDN, which eliminates close false positive proposals successively. With the above characteristics, MSR2N is more suitable and robust for ship detection in SAR images. The experimental results on SAR ship detection dataset (SSDD) show that the MSR2N has achieved state-of-the-art performance.

This paper proposes a novel method of semantic segmentation, consisting of modified dilated residual network, atrous pyramid pooling module, and backpropagation, that is applicable to augmented reality (AR). In the proposed method, the modified dilated residual network extracts a feature map from the original images and maintains spatial information. The atrous pyramid pooling module places convolutions in parallel and layers feature maps in a pyramid shape to extract objects occupying small areas in the image, these are converted into one channel using a 1 &times, 1 convolution. Backpropagation compares the semantic segmentation obtained through convolution from the final feature map with the ground truth provided by a database. Losses can be reduced by applying backpropagation to the modified dilated residual network to change the weighting. The proposed method was compared with other methods on the Cityscapes and PASCAL VOC 2012 databases. The proposed method achieved accuracies of 82.8 and 89.8 mean intersection over union (mIOU) and frame rates of 61 and 64.3 frames per second (fps) for the Cityscapes and PASCAL VOC 2012 databases, respectively. These results prove the applicability of the proposed method for implementing natural AR applications at actual speeds because the frame rate is greater than 60 fps.

Although unsupervised representation learning (RL) can tackle the performance deterioration caused by limited labeled data in synthetic aperture radar (SAR) object classification, the neglected discriminative detailed information and the ignored distinctive characteristics of SAR images can lead to performance degradation. In this paper, an unsupervised multi-scale convolution auto-encoder (MSCAE) was proposed which can simultaneously obtain the global features and local characteristics of targets with its U-shaped architecture and pyramid pooling modules (PPMs). The compact depth-wise separable convolution and the deconvolution counterpart were devised to decrease the trainable parameters. The PPM and the multi-scale feature learning scheme were designed to learn multi-scale features. Prior knowledge of SAR speckle was also embedded in the model. The reconstruction loss of the MSCAE was measured by the structural similarity index metric (SSIM) of the reconstructed data and the images filtered by the improved Lee sigma filter. A speckle suppression restriction was also added in the objective function to guarantee that the speckle suppression procedure would take place in the feature learning stage. Experimental results with the MSTAR dataset under the standard operating condition and several extended operating conditions demonstrated the effectiveness of the proposed model in SAR object classification tasks.

With the rapid development of flexible vision sensors and visual sensor networks, computer vision tasks, such as object detection and tracking, are entering a new phase. Accordingly, the more challenging comprehensive task, including instance segmentation, can develop rapidly. Most state-of-the-art network frameworks, for instance, segmentation, are based on Mask R-CNN (mask region-convolutional neural network). However, the experimental results confirm that Mask R-CNN does not always successfully predict instance details. The scale-invariant fully convolutional network structure of Mask R-CNN ignores the difference in spatial information between receptive fields of different sizes. A large-scale receptive field focuses more on detailed information, whereas a small-scale receptive field focuses more on semantic information. So the network cannot consider the relationship between the pixels at the object edge, and these pixels will be misclassified. To overcome this problem, Mask-Refined R-CNN (MR R-CNN) is proposed, in which the stride of ROIAlign (region of interest align) is adjusted. In addition, the original fully convolutional layer is replaced with a new semantic segmentation layer that realizes feature fusion by constructing a feature pyramid network and summing the forward and backward transmissions of feature maps of the same resolution. The segmentation accuracy is substantially improved by combining the feature layers that focus on the global and detailed information. The experimental results on the COCO (Common Objects in Context) and Cityscapes datasets demonstrate that the segmentation accuracy of MR R-CNN is about 2% higher than that of Mask R-CNN using the same backbone. The average precision of large instances reaches 56.6%, which is higher than those of all state-of-the-art methods. In addition, the proposed method requires low time cost and is easily implemented. The experiments on the Cityscapes dataset also prove that the proposed method has great generalization ability.

Object detection in point cloud data is one of the key components in computer vision systems, especially for autonomous driving applications. In this work, we present Voxel-Feature Pyramid Network, a novel one-stage 3D object detector that utilizes raw data from LIDAR sensors only. The core framework consists of an encoder network and a corresponding decoder followed by a region proposal network. Encoder extracts and fuses multi-scale voxel information in a bottom-up manner, whereas decoder fuses multiple feature maps from various scales by Feature Pyramid Network in a top-down way. Extensive experiments show that the proposed method has better performance on extracting features from point data and demonstrates its superiority over some baselines on the challenging KITTI-3D benchmark, obtaining good performance on both speed and accuracy in real-world scenarios.

Nowadays, the concept of Industry 4.0 aims to improve factories’ competitiveness. Usually, manufacturing production is guided by standards to segment and distribute its processes and implementations. However, industry 4.0 requires innovative proposals for disruptive technologies that engage the entire production process in factories, not just a partial improvement. One of these disruptive technologies is the Digital Twin (DT). This advanced virtual model runs in real-time and can predict, detect, and classify normal and abnormal operating conditions in factory processes. The Automation Pyramid (AP) is a conceptual element that enables the efficient distribution and connection of different actuators in enterprises, from the shop floor to the decision-making levels. When a DT is deployed into a manufacturing system, generally, the DT focuses on the low-level that is named field level, which includes the physical devices such as controllers, sensors, and so on. Thus, the partial automation based on the DT is accomplished, and the information between all manufacturing stages could be decremented. Hence, to achieve a complete improvement of the manufacturing system, all the automation pyramid levels must be included in the DT concept. An artificial intelligent management system could create an interconnection between them that can manage the information. As a result, this paper proposed a complete DT structure covering all automation pyramid stages using Artificial Intelligence (AI) to model each stage of the AP based on the Digital Twin concept. This work proposes a virtual model for each level of the traditional AP and the interactions among them to flow and control information efficiently. Therefore, the proposed model is a valuable tool in improving all levels of an industrial process. In addition, It is presented a case study where the DT concept for modular workstations underpins the development of technologies within the framework of the Automation Pyramid model is implemented into a didactic manufacturing system.

Deep learning-based aircraft detection methods have been increasingly implemented in recent years. However, due to the multi-resolution imaging modes, aircrafts in different images show very wide diversity on size, view and other visual features, which brings great challenges to detection. Although standard deep convolution neural networks (DCNN) can extract rich semantic features, they destroy the bottom-level location information. The features of small targets may also be submerged by redundant top-level features, resulting in poor detection. To address these problems, we proposed a compact multi-scale dense convolutional neural network (MS-DenseNet) for aircraft detection in remote sensing images. Herein, DenseNet was utilized for feature extraction, which enhances the propagation and reuse of the bottom-level high-resolution features. Subsequently, we combined feature pyramid network (FPN) with DenseNet to form a MS-DenseNet for learning multi-scale features, especially features of small objects. Finally, by compressing some of the unnecessary convolution layers of each dense block, we designed three new compact architectures: MS-DenseNet-41, MS-DenseNet-65, and MS-DenseNet-77. Comparative experiments showed that the compact MS-DenseNet-65 obtained a noticeable improvement in detecting small aircrafts and achieved state-of-the-art performance with a recall of 94% and an F1-score of 92.7% and cost less computational time. Furthermore, the experimental results on robustness of UCAS-AOD and RSOD datasets also indicate the good transferability of our method.

Outdoor scene understanding based on the results of point cloud classification plays an important role in mobile robots and autonomous vehicles equipped with a light detection and ranging (LiDAR) system. In this paper, a novel model named Panoramic Bearing Angle (PBA) images is proposed which is generated from 3D point clouds. In a PBA model, laser point clouds are projected onto the spherical surface to establish the correspondence relationship between the laser ranging point and the image pixels, and then we use the relative location relationship of the laser point in the 3D space to calculate the gray value of the corresponding pixel. To extract robust features from 3D laser point clouds, both image pyramid model and point cloud pyramid model are utilized to extract multiple-scale features from PBA images and original point clouds, respectively. A Random Forest classifier is used to accomplish feature screening on extracted high-dimensional features to obtain the initial classification results. Moreover, reclassification is carried out to correct the misclassification points by remapping the classification results into the PBA images and using superpixel segmentation, which makes full use of the contextual information between laser points. Within each superpixel block, the reclassification is carried out again based on the results of the initial classification results, so as to correct some misclassification points and improve the classification accuracy. Two datasets published by ETH Zurich and MINES ParisTech are used to test the classification performance, and the results show the precision and recall rate of the proposed algorithms.

Vehicle Logo Recognition (VLR) is an important part of vehicle behavior analysis and can provide supplementary information for vehicle identification, which is an essential research topic in robotic systems. However, the inaccurate extraction of vehicle logo candidate regions will affect the accuracy of logo recognition. Additionally, the existing methods have low recognition rate for most small vehicle logos and poor performance under complicated environments. A VLR method based on enhanced matching, constrained region extraction and SSFPD network is proposed in this paper to solve the aforementioned problems. A constrained region extraction method based on segmentation of the car head and car tail is proposed to accurately extract the candidate region of logo. An enhanced matching method is proposed to improve the detection performance of small objects, which augment each of training images by copy-pasting small objects many times in the unconstrained region. A single deep neural network based on a reduced ResNeXt model and Feature Pyramid Networks is proposed in this paper, which is named as Single Shot Feature Pyramid Detector (SSFPD). The SSFPD uses the reduced ResNeXt to improve classification performance of the network and retain more detailed information for small-sized vehicle logo detection. Additionally, it uses the Feature Pyramid Networks module to bring in more semantic context information to build several high-level semantic feature maps, which effectively improves recognition performance. Extensive evaluations have been made on self-collected and public vehicle logo datasets. The proposed method achieved 93.79% accuracy on the Common Vehicle Logos Dataset and 99.52% accuracy on another public dataset, respectively, outperforming the existing methods.

To improve the fault diagnosis performance for rotating machinery, an efficient, noise-resistant end-to-end deep learning (DL) algorithm is proposed based on the advantages of the wavelet packet transform in vibration signal processing (the capability to extract multiscale information and more spectral distribution features) and deep convolutional neural networks (good classification performance, data-driven design and high transfer-learning ability). First, a vibration signal is subjected to pyramid wavelet packet decomposition, and each sub-band coefficient is used as the input for each channel of a deep convolutional network (DCN). Then, based on the lightweight modeling requirements and techniques, a new DCN structure is designed for the fault diagnosis. The proposed algorithm is compared with the support vector machine algorithm and the published DL algorithms based on a bearing dataset produced by Case Western Reserve University. The experimental results show that the proposed algorithm is superior to the existing algorithms in terms of accuracy, memory space, computational complexity, noise resistance, and transfer performance, producing good results.

Vehicle detection with category inference on video sequence data is an important but challenging task for urban traffic surveillance. The difficulty of this task lies in the fact that it requires accurate localization of relatively small vehicles in complex scenes and expects real-time detection. In this paper, we present a vehicle detection framework that improves the performance of the conventional Single Shot MultiBox Detector (SSD), which effectively detects different types of vehicles in real-time. Our approach, which proposes the use of different feature extractors for localization and classification tasks in a single network, and to enhance these two feature extractors through deconvolution (D) and pooling (P) between layers in the feature pyramid, is denoted as DP-SSD. In addition, we extend the scope of the default box by adjusting its scale so that smaller default boxes can be exploited to guide DP-SSD training. Experimental results on the UA-DETRAC and KITTI datasets demonstrate that DP-SSD can achieve efficient vehicle detection for real-world traffic surveillance data in real-time. For the UA-DETRAC test set trained with UA-DETRAC trainval set, DP-SSD with the input size of 300 &times, 300 achieves 75.43% mAP (mean average precision) at the speed of 50.47 FPS (frames per second), and the framework with a 512 &times, 512 sized input reaches 77.94% mAP at 25.12 FPS using an NVIDIA GeForce GTX 1080Ti GPU. The DP-SSD shows comparable accuracy, which is better than those of the compared state-of-the-art models, except for YOLOv3.

Building extraction from high spatial resolution remote sensing images is a hot spot in the field of remote sensing applications and computer vision. This paper presents a semantic segmentation model, which is a supervised method, named Pyramid Self-Attention Network (PISANet). Its structure is simple, because it contains only two parts: one is the backbone of the network, which is used to learn the local features (short distance context information around the pixel) of buildings from the image, the other part is the pyramid self-attention module, which is used to obtain the global features (long distance context information with other pixels in the image) and the comprehensive features (includes color, texture, geometric and high-level semantic feature) of the building. The network is an end-to-end approach. In the training stage, the input is the remote sensing image and corresponding label, and the output is probability map (the probability that each pixel is or is not building). In the prediction stage, the input is the remote sensing image, and the output is the extraction result of the building. The complexity of the network structure was reduced so that it is easy to implement. The proposed PISANet was tested on two datasets. The result shows that the overall accuracy reached 94.50 and 96.15%, the intersection-over-union reached 77.45 and 87.97%, and F1 index reached 87.27 and 93.55%, respectively. In experiments on different datasets, PISANet obtained high overall accuracy, low error rate and improved integrity of individual buildings.

Accurate and robust detection of multi-class objects in very high resolution (VHR) aerial images has been playing a significant role in many real-world applications. The traditional detection methods have made remarkable progresses with horizontal bounding boxes (HBBs) due to CNNs. However, HBB detection methods still exhibit limitations including the missed detection and the redundant detection regions, especially for densely-distributed and strip-like objects. Besides, large scale variations and diverse background also bring in many challenges. Aiming to address these problems, an effective region-based object detection framework named Multi-scale Feature Integration Attention Rotation Network (MFIAR-Net) is proposed for aerial images with oriented bounding boxes (OBBs), which promotes the integration of the inherent multi-scale pyramid features to generate a discriminative feature map. Meanwhile, the double-path feature attention network supervised by the mask information of ground truth is introduced to guide the network to focus on object regions and suppress the irrelevant noise. To boost the rotation regression and classification performance, we present a robust Rotation Detection Network, which can generate efficient OBB representation. Extensive experiments and comprehensive evaluations on two publicly available datasets demonstrate the effectiveness of the proposed framework.

Generic object detection algorithms for natural images have been proven to have excellent performance. In this paper, fabric defect detection on optical image datasets is systematically studied. In contrast to generic datasets, defect images are multi-scale, noise-filled, and blurred. Back-light intensity would also be sensitive for visual perception. Large-scale fabric defect datasets are collected, selected, and employed to fulfill the requirements of detection in industrial practice in order to address these imbalanced issues. An improved two-stage defect detector is constructed for achieving better generalization. Stacked feature pyramid networks are set up to aggregate cross-scale defect patterns on interpolating mixed depth-wise block in stage one. By sharing feature maps, center-ness and shape branches merges cascaded modules with deformable convolution to filter and refine the proposed guided anchors. After balanced sampling, the proposals are down-sampled by position-sensitive pooling for region of interest, in order to characterize interactions among fabric defect images in stage two. The experiments show that the end-to-end architecture improves the occluded defect performance of region-based object detectors as compared with the current detectors.

Urban land cover and land use mapping plays an important role in urban planning and management. In this paper, novel multi-scale deep learning models, namely ASPP-Unet and ResASPP-Unet are proposed for urban land cover classification based on very high resolution (VHR) satellite imagery. The proposed ASPP-Unet model consists of a contracting path which extracts the high-level features, and an expansive path, which up-samples the features to create a high-resolution output. The atrous spatial pyramid pooling (ASPP) technique is utilized in the bottom layer in order to incorporate multi-scale deep features into a discriminative feature. The ResASPP-Unet model further improves the architecture by replacing each layer with residual unit. The models were trained and tested based on WorldView-2 (WV2) and WorldView-3 (WV3) imageries over the city of Beijing. Model parameters including layer depth and the number of initial feature maps (IFMs) as well as the input image bands were evaluated in terms of their impact on the model performances. It is shown that the ResASPP-Unet model with 11 layers and 64 IFMs based on 8-band WV2 imagery produced the highest classification accuracy (87.1% for WV2 imagery and 84.0% for WV3 imagery). The ASPP-Unet model with the same parameter setting produced slightly lower accuracy, with overall accuracy of 85.2% for WV2 imagery and 83.2% for WV3 imagery. Overall, the proposed models outperformed the state-of-the-art models, e.g., U-Net, convolutional neural network (CNN) and Support Vector Machine (SVM) model over both WV2 and WV3 images, and yielded robust and efficient urban land cover classification results.

Object detection in very high-resolution (VHR) aerial images is an essential step for a wide range of applications such as military applications, urban planning, and environmental management. Still, it is a challenging task due to the different scales and appearances of the objects. On the other hand, object detection task in VHR aerial images has improved remarkably in recent years due to the achieved advances in convolution neural networks (CNN). Most of the proposed methods depend on a two-stage approach, namely: a region proposal stage and a classification stage such as Faster R-CNN. Even though two-stage approaches outperform the traditional methods, their optimization is not easy and they are not suitable for real-time applications. In this paper, a uniform one-stage model for object detection in VHR aerial images has been proposed. In order to tackle the challenge of different scales, a densely connected feature pyramid network has been proposed by which high-level multi-scale semantic feature maps with high-quality information are prepared for object detection. This work has been evaluated on two publicly available datasets and outperformed the current state-of-the-art results on both in terms of mean average precision (mAP) and computation time.

Autonomous harvesting shows a promising prospect in the future development of theagriculture industry, while the vision system is one of the most challenging components in theautonomous harvesting technologies. This work proposes a multi-function network to perform thereal-time detection and semantic segmentation of apples and branches in orchard environments byusing the visual sensor. The developed detection and segmentation network utilises the atrous spatialpyramid pooling and the gate feature pyramid network to enhance feature extraction ability of thenetwork. To improve the real-time computation performance of the network model, a lightweightbackbone network based on the residual network architecture is developed. From the experimentalresults, the detection and segmentation network with ResNet-101 backbone outperformed on thedetection and segmentation tasks, achieving an F1 score of 0.832 on the detection of apples and 87.6%and 77.2% on the semantic segmentation of apples and branches, respectively. The network modelwith lightweight backbone showed the best computation efficiency in the results. It achieved an F1score of 0.827 on the detection of apples and 86.5% and 75.7% on the segmentation of apples andbranches, respectively. The weights size and computation time of the network model with lightweightbackbone were 12.8 M and 32 ms, respectively. The experimental results show that the detection andsegmentation network can effectively perform the real-time detection and segmentation of applesand branches in orchards.

Human action recognition is an increasingly important research topic in the fields of video sensing, analysis and understanding. Caused by unconstrained sensing conditions, there exist large intra-class variations and inter-class ambiguities in realistic videos, which hinder the improvement of recognition performance for recent vision-based action recognition systems. In this paper, we propose a generalized pyramid matching kernel (GPMK) for recognizing human actions in realistic videos, based on a multi-channel “bag of words” representation constructed from local spatial-temporal features of video clips. As an extension to the spatial-temporal pyramid matching (STPM) kernel, the GPMK leverages heterogeneous visual cues in multiple feature descriptor types and spatial-temporal grid granularity levels, to build a valid similarity metric between two video clips for kernel-based classification. Instead of the predefined and fixed weights used in STPM, we present a simple, yet effective, method to compute adaptive channel weights of GPMK based on the kernel target alignment from training data. It incorporates prior knowledge and the data-driven information of different channels in a principled way. The experimental results on three challenging video datasets (i.e., Hollywood2, Youtube and HMDB51) validate the superiority of our GPMK w.r.t. the traditional STPM kernel for realistic human action recognition and outperform the state-of-the-art results in the literature.

The rapid development of remote sensing (RS) technology has resulted in the proliferation of high-resolution images. There are challenges involved in not only storing large volumes of RS images but also in rapidly retrieving the images for ocean disaster analysis such as for storm surges and typhoon warnings. In this paper, we present an efficient retrieval of massive ocean RS images via a Cloud-based mean-shift algorithm. Distributed construction method via the pyramid model is proposed based on the maximum hierarchical layer algorithm and used to realize efficient storage structure of RS images on the Cloud platform. We achieve high-performance processing of massive RS images in the Hadoop system. Based on the pyramid Hadoop distributed file system (HDFS) storage method, an improved mean-shift algorithm for RS image retrieval is presented by fusion with the canopy algorithm via Hadoop MapReduce programming. The results show that the new method can achieve better performance for data storage than HDFS alone and WebGIS-based HDFS. Speedup and scaleup are very close to linear changes with an increase of RS images, which proves that image retrieval using our method is efficient.