Small Target Size Research Articles

Abstract 1874: Use of high throughput SPR to characterize antibodies to common cancer biomarkers

Abstract As a disease of mis-regulation, cancer is characterized both genetically and proteomically, requiring high quality tools to dissect the change difference between cell states. Among the reagents used to characterize biological systems at the protein level, antibodies are routinely applied, and engineered versions have proven useful for treating multiple therapeutic indications. We have generated a human proteome wide toolbox of polyclonal antibodies, identifying three antigenic 14mer sequences within each of the 20,000+ open reading frames at N-terminal, middle and C-terminal regions. Using a high throughput SPR system, Carterra LSA, we are able to define average on-rates, off-rates, and KD’s for pAb’s generated to important cancer regulators and biomarkers, including KRAS, STAT3, and Gli1. Due to the small size of the peptide targets, the kinetics of the observed interactions with polyclonal antibodies exhibit remarkable 1:1 fit typically observed when measuring monoclonal antibody affinities. The platform is a sensitive system capable of both identifying multiple antibody-antigen interactions with KD in the nanomolar to picomolar range, but also as a mass screening tool for developing de novo antibody pairs. We define the kinetic characteristics of these antibody reagents relative to performance in western blotting specifically. Improving the characterization of such widely used tools is critical for advancing translational opportunities into use in the clinic. Citation Format: Sergei Bibikov, John Rosenfeld, Kevin J. Harvey. Use of high throughput SPR to characterize antibodies to common cancer biomarkers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1874.

Infrared Bird Target Detection Based on Temporal Variation Filtering and a Gaussian Heat-Map Perception Network

Flying bird detection has recently attracted increasing attention in computer vision. However, compared to conventional object detection tasks, it is much more challenging to trap flying birds in infrared videos due to small target size, complex backgrounds, and dim shapes. In order to solve the problem of poor detection performance caused by insufficient feature information of small and dim birds, this paper suggests a method of detecting birds in outdoor environments using image pre-processing and deep learning, called temporal Variation filtering (TVF) and Gaussian heatmap perception network (GHPNet), respectively. TVF separates the dynamic background from moving creatures. Using bird appearance features that are brightest at the center and gradually darker outwards, the size-adaptive Gaussian kernel is used to generate the ground truth of the region of interest (ROI). In order to fuse the features from different scales and to highlight the saliency of the target, the GHPNet network integrates VGG-16 and maximum-no-pooling filterer into a U-Net network. The comparative experiments demonstrate that the proposed method outperforms those that are state-of-the-art in detecting bird targets in real-world infrared images.

Detection algorithm of aerial vehicle target based on improved YOLOv3

In order to solve the problems of large number, small size and low detection accuracy of vehicle targets in aerial photography, a target detection algorithm based on improved YOLOv3 is proposed in this paper. Firstly, aiming at the problem of small target information loss of road vehicles, a new detection size is added.Secondly, in order to better detect small targets, a 104x104 scale detection layer is added on the basis of the three detection layers of the traditional yolov3 network structure.The k-means + + algorithm is used to cluster the data set, and a new ground target detection candidate frame is obtained. The loss function is improved by using Focal loss function in the classification loss function and using DIoUloss function based on IoUloss improvement in the regression loss function.The improved YOLOv3 algorithm can identify the road vehicle target more accurately without the decline of speed, and reduce the miss rate. The improved algorithm is tested on Visdrone dataset, and the experimental data show that the average accuracy of the improved algorithm is 94.04%, and the average detection accuracy (mAP) is improved by 2.94%.The detection accuracy of the proposed improved YOLOv3 algorithm is better than that of YOLOv3.

Dosimetric dependencies on target geometry and size in radioiodine therapy for differentiated thyroid cancer

Purpose Radioiodine therapy is used in most disease stages for differentiated thyroid cancer. Its success depends on several factors, such as lesion size, completeness of surgery, extent of metastasis and tumoural iodine avidity. We aimed to investigate the importance of non-spherical geometries and size of metastases and thyroid remnants for the absorbed dose delivered. Methods Absorbed doses and energy depositions from homogeneously distributed iodine-131 in clinically relevant geometries and sizes were calculated using Monte Carlo simulations with MCNP6. A total of 162 volumes with different sizes and geometries corresponding to spheres, and prolate or oblate spheroids were simulated. Results Oblate and prolate spheroids had worse radiation coverage compared to spheres for equal masses, up to a difference of 38% for the most eccentric oblate spheroids and smallest masses simulated (a micrometastasis of mass 0.005 g). The differences in coverage could be explained by different volume - to - surface - area ratios of the spheroids. The impact of size alone caused up to 71% lower absorbed doses per decay in a spherical target mass of 0.005 g compared to 50 g. Conclusions While the iodine avidity, and therefore the total amount of decays, is the predominant contributing factor to absorbed dose in radioiodine therapy, eccentric spheroids and small target sizes can receive substantially lower absorbed doses from the same administration of radioiodine.

Rescaling-Assisted Super-Resolution for Medium-Low Resolution Remote Sensing Ship Detection

Medium-low resolution (M-LR) remote sensing ship detection is a challenging problem due to the small target sizes and insufficient appearance information. Although image super resolution (SR) has become a popular solution in recent years, the ability of image SR is limited since much information is lost in input images. Inspired by the powerful information embedding ability of the encoder in image rescaling, in this paper, we introduce image rescaling to guide the training of image SR. Specifically, we add an adaption module before the SR network, and use the pre-trained rescaling network to guide the optimization of the adaption module. In this way, more information is embedded in the adapted M-LR images, and the subsequent SR module can utilize more information to achieve better performance. Extensive experimental results demonstrate the effectiveness of our method on image SR. More importantly, our method can be used as a pre-processing approach to improve the detection performance.

An Efficient Lightweight SAR Ship Target Detection Network with Improved Regression Loss Function and Enhanced Feature Information Expression.

It is difficult to identify the ship images obtained by a synthetic aperture radar (SAR) due to the influence of dense ships, complex background and small target size, so a deep learning-based target detection algorithm was introduced to obtain better detection performance. However, in order to achieve excellent performance, most of the current target detection algorithms focus on building deep and high-width neural networks, resulting in bloated network structure and reduced detection speed, which is not conducive to the practical application of target detection algorithms. Thereby, an efficient lightweight network Efficient-YOLO for ship detection in complex situations is proposed in the present work. Firstly, a new regression loss function ECIOU is proposed to enhance the detection boxes localization accuracy and model convergence speed. Secondly, We propose the SCUPA module to enhance the multiplexing of picture feature information and the model generalization performance. Thirdly, The GCHE module is proposed to strengthen the network’s ability to extract feature information. At last, the effectiveness of our method is tested on the specialized ship dataset: SSDD and HRSID datasets. The results show that Efficient-YOLO outperforms other state-of-the-art algorithms in accuracy, recall and detection speed, with smaller model complexity and model size.

Human Detection in Aerial Thermal Images Using Faster R-CNN and SSD Algorithms

The automatic detection of humans in aerial thermal imagery plays a significant role in various real-time applications, such as surveillance, search and rescue and border monitoring. Small target size, low resolution, occlusion, pose, and scale variations are the significant challenges in aerial thermal images that cause poor performance for various state-of-the-art object detection algorithms. Though many deep-learning-based object detection algorithms have shown impressive performance for generic object detection tasks, their ability to detect smaller objects in the aerial thermal images is analyzed through this study. This work carried out the performance evaluation of Faster R-CNN and single-shot multi-box detector (SSD) algorithms with different backbone networks to detect human targets in aerial view thermal images. For this purpose, two standard aerial thermal datasets having human objects of varying scale are considered with different backbone networks, such as ResNet50, Inception-v2, and MobileNet-v1. The evaluation results demonstrate that the Faster R-CNN model trained with the ResNet50 network architecture out-performed in terms of detection accuracy, with a mean average precision (mAP at 0.5 IoU) of 100% and 55.7% for the test data of the OSU thermal dataset and AAU PD T datasets, respectively. SSD with MobileNet-v1 achieved the highest detection speed of 44 frames per second (FPS) on the NVIDIA GeForce GTX 1080 GPU. Fine-tuning the anchor parameters of the Faster R-CNN ResNet50 and SSD Inception-v2 algorithms caused remarkable improvement in mAP by 10% and 3.5%, respectively, for the challenging AAU PD T dataset. The experimental results demonstrated the application of Faster R-CNN and SSD algorithms for human detection in aerial view thermal images, and the impact of varying backbone network and anchor parameters on the performance improvement of these algorithms.

Aircraft Detection in Remote Sensing Imagery Based on Improved YOLOv4

Abstract A method is provided to resolve the issues of false recognition and lower recall of the you only look once version 4 (YOLOv4) algorithm in the aircraft targets prediction of remote sensing imagery. The method is based on the YOLOv4. According to the characteristics of small size and intensive arrangement of aircraft targets in remote sensing images, the backbone is mended to boost the prediction capacity of smaller targets. To strengthen the transmission and reuse of network features, the connection mode of the residual module in YOLOv4 is changed to the dense connection mode, and the five residual blocks in the backbone network are modified to the corresponding dense connection blocks to reduce the network layers and model redundancy. The experiment results reveal that compared with the YOLOv4, our method can enlarge targets features in smaller image regions, thereby availably increasing the prediction precision of aircraft targets.

Classification of Underwater Target Based on S-ResNet and Modified DCGAN Models.

Underwater target classification has been an important topic driven by its general applications. Convolutional neural network (CNN) has been shown to exhibit excellent performance on classifications especially in the field of image processing. However, when applying CNN and related deep learning models to underwater target classifications, the problems, including small sample size of underwater target and low complexity requirement, impose a great challenge. In this paper, we have proposed the modified DCGAN model to augment data for targets with small sample size. The data generated from the proposed model help to improve classification performance under imbalanced category conditions. Furthermore, we have proposed the S-ResNet model to obtain good classification accuracy while significantly reducing complexity of the model, and achieve a good tradeoff between classification accuracy and model complexity. The effectiveness of proposed models is verified through measured data from sea trial and lake tests.

SDWBF Algorithm: A Novel Pedestrian Detection Algorithm in the Aerial Scene

Due to the large amount of video data from UAV aerial photography and the small target size from the aerial perspective, pedestrian detection in drone videos remains a challenge. To detect objects in UAV images quickly and accurately, a small-sized pedestrian detection algorithm based on the weighted fusion of static and dynamic bounding boxes is proposed. First, a weighted filtration algorithm for redundant frames was applied using the inter-frame pixel difference algorithm cascading vision and structural similarity, which solved the redundancy of the UAV video data, thereby reducing the delay. Second, the pre-training and detector learning datasets were scale matched to address the feature representation loss caused by the scale mismatch between datasets. Finally, the static bounding extracted by YOLOv4 and the motion bounding boxes extracted by LiteFlowNet were subject to the weighted fusion algorithm to enhance the semantic information and solve the problem of missing and multiple detections in UAV object detection. The experimental results showed that the small object recognition method proposed in this paper enabled reaching an mAP of 70.91% and an IoU of 57.53%, which were 3.51% and 2.05% higher than the mainstream target detection algorithm.

Technical note: Toward implementation of MR‐guided radiation therapy for laryngeal cancer with healthy volunteer imaging and a custom MR‐CT larynx phantom

Internal motion of the larynx can cause normal tissue toxicity and/or tumor underdosage during radiotherapy. MR-guided radiation therapy (MRgRT) provides improved soft-tissue contrast for patient setup and real-time gating of radiation based on cine imaging of tumor motion, potentially making it an advantageous modality for laryngeal treatments. However, there are potential concerns regarding the small target size, proximity to heterogeneous tissue interfaces in the airway that may cause dosimetric errors in the presence of the magnetic field, and uncertainty about the ability of MR-linear accelerator (MR-Linac) systems to visualize and track laryngeal motion. To date, there have been no reports of the use of MRgRT for laryngeal treatments. A healthy volunteer was imaged on a ViewRay MRIdian MR-Linac. Organs-at-risk and a laryngeal pseudo target were contoured and used to generate a stereotactic body radiotherapy plan. A custom phantom was created using 3D-printing based on structures delineated on the volunteer images to construct an enclosure containing the target and airway anatomy, with a gap for radiochromic film, and filled with gelatin . The treatment plan was mapped onto the phantom and delivered dose assessed on radiochromic film with global normalization and a 10% dose threshold. A cine MR of the volunteer was acquired to assess the magnitude of larynx motion with speaking and swallowing, and system's ability to gate radiation. A clinically acceptable laryngeal treatment plan and larynx phantom that was MR and computed tomography-visible were successfully created. The delivered dose had good agreement with the treatment plan with a gamma passing rate of 96.5% (3%/2mm). The MR-Linac was able to visualize, track, and gate larynx motion. The MRgRT workflow for laryngeal treatments was assessed and performed in preparation for clinical implementation on the MR-Linac, demonstrating that it is feasible to treat laryngeal cancer patients on the MR-Linac.

Spatio-Temporal Dual-Branch Network With Predictive Feature Learning for Satellite Video Object Segmentation

Satellite video is an important new earth observation data source that can be used to acquire large-scale dynamic information. Satellite video object segmentation (SVOS) is aimed at separating the foreground and background of satellite video and is a fundamental processing task for satellite video. To date, the state-of-the-art research into SVOS has mainly focused on unsupervised target extraction methods through hand-crafted features and post-processing operations, which are prone to obtaining incomplete contours of the targets and result in foreground aperture problem. Furthermore, the small size of targets and appearance deformation make the SVOS more difficult. Therefore, in this article, a spatio-temporal dual-branch network is proposed with predictive feature learning for the SVOS task. The proposed model consists of a temporal coherence branch and a spatial segmentation branch. In the temporal coherence branch, the Wasserstein generative adversarial network (WGAN) architecture is utilized for the future frame prediction to exploit temporal information, which captures the dynamic appearance and motion cues from the unlabeled satellite video data through predictive feature learning module in an adversarial manner. As a result, the proposed method can obtain segmentation results with temporal consistency, while avoiding the generation of optical flow images. In the spatial segmentation branch, a fully convolutional network (FCN) is used to extract the high-level spatial information of the satellite video and achieve end-to-end SVOS, without any post-processing operations. In the network implementation, boundary loss is used to solve the highly unbalanced segmentation problem caused by small size of the targets. The two branches of the network are also mutually constrained, to improve the final object segmentation results. The visual and quantitative results of three experiments all demonstrate that the proposed method outperforms the other current SVOS models.

A High-Effective Implementation of Ship Detector for SAR Images

Synthetic aperture radar (SAR) can be applied to observe the sea surface and detect ship targets. Images obtained by SAR can be hard to read because of the denseness of ships, extremely unbalanced foreground–background, and small target size. The existing objective detection approaches achieve superior performance by sacrificing detection speed and flexibility on computational resource due to redundant model parameters. These features happens to be extremely crucial for SAR ship detection in real-time scenarios. To effectively tradeoff the issue on both time and space complexity, we propose a novel objective detection method tailored for SAR ship detection problem. We examine the effectiveness of our method on two open datasets: high resolution SAR images dataset (HRSID) and large-scale SAR ship detection dataset-v1.0 (LS-SSDD-V1.0). The results show better performance over the state-of-the-art You Only Look Once version 4 (YOLOv4) framework in terms of accuracy, efficiency, and model complexity.

Lightweight tea bud recognition network integrating GhostNet and YOLOv5.

Aiming at the problems of low detection accuracy and slow speed caused by the complex background of tea sprouts and the small target size, this paper proposes a tea bud detection algorithm integrating GhostNet and YOLOv5. To reduce parameters, the GhostNet module is specially introduced to shorten the detection speed. A coordinated attention mechanism is then added to the backbone layer to enhance the feature extraction ability of the model. A bi-directional feature pyramid network (BiFPN) is used in the neck layer of feature fusion to increase the fusion between shallow and deep networks to improve the detection accuracy of small objects. Efficient intersection over union (EIOU) is used as a localization loss to improve the detection accuracy in the end. The experimental results show that the precision of GhostNet-YOLOv5 is 76.31%, which is 1.31, 4.83, and 3.59% higher than that of Faster RCNN, YOLOv5 and YOLOv5-Lite respectively. By comparing the actual detection effects of GhostNet-YOLOv5 and YOLOv5 algorithm on buds in different quantities, different shooting angles, and different illumination angles, and taking F1 score as the evaluation value, the results show that GhostNet-YOLOv5 is 7.84, 2.88, and 3.81% higher than YOLOv5 algorithm in these three different environments.

YOLOv5-based Defect Detection Model for Hot Rolled Strip Steel

In the defect detection of hot-rolled strip steel, there are often problems of too small target size and unclear features that lead to wrong detection and missed detection, for which a YOLOv5-based defect detection method for hot-rolled strip steel is proposed in this paper. Firstly, the overall architecture of the method is proposed, and then the algorithm implementation process is highlighted. Experimental analysis shows that the average detection accuracy using YOLOv5 is improved by 11.9% compared to YOLOv4 improved, with stronger generalization capability, faster detection speed, and lower error and miss detection rate.

Chapter 7 - Neurophysiology during movement disorder surgery

During stereotactic procedures for treating medically refractory movement disorders, intraoperative neurophysiology shifts its focus from simply monitoring the effects of surgery to an integral part of the surgical procedure. The small size, poor visualization, and physiologic nature of these deep brain targets compel the surgeon to rely on some form of physiologic for confirmation of proper anatomic targeting. Even given the newer reliance on imaging and asleep deep brain stimulator electrode placement, it is still a physiologic target and thus some form of intraoperative physiology is necessary. This chapter reviews the neurophysiologic monitoring method of microelectrode recording that is commonly employed during these neurosurgical procedures today.

DSFNet: Dynamic and Static Fusion Network for Moving Object Detection in Satellite Videos

Moving object detection (MOD) in satellite videos remains challenging due to the extremely small size of the interested targets and the highly complex background. Both the intra-frame (static) and inter-frame (dynamic) information are of great importance to MOD. In this letter, we propose a two-stream detection network named dynamic and static fusion network (DSFNet) to tackle the MOD problem in satellite videos. Specifically, the DSFNet is composed of a 2-D backbone to extract static context information from a single frame and a lightweight 3-D backbone to extract dynamic motion cues from consecutive frames. Then the extracted static and dynamic features are fused and fed into the detection head to detect the moving targets in satellite videos. We conduct extensive experiments on videos collected from Jilin-1 satellite and the results have demonstrated the effectiveness and robustness of the proposed DSFNet. Experimental results show that our DSFNet achieves the-state-of-the-art performance.

Cross-Connected Bidirectional Pyramid Network for Infrared Small-Dim Target Detection

Infrared small-dim target detection is an important technology in the fields of infrared guidance, anti-missile, and tracking system. Due to the small size of targets, no obvious structure information, and low image signal-to-noise ratio (SNR), infrared small-dim target detection is still a challenging task. In this letter, a cross-connected bidirectional pyramid network (CBP-Net) is proposed for infrared small-dim target detection. The main body of the CBP-Net is to embed a bottom-up pyramid in the feature pyramid network (FPN), which is designed to provide more comprehensive target information by connecting with the original multi-scale features and the top-down pyramid. The bottom-up pyramid together with the top-down pyramid forms the proposed bidirectional pyramid structure. Then, an region of interest (ROI) feature augment module (RFA) composed of deformable ROI pooling and position attention is designed to fuse multi-scale ROI features and enhance the spatial information of the small-dim target. Besides, a regular constraint loss (RCL) is introduced to restrict multi-scale feature fusion to learn more precise target location information. Experimental results on two challenging datasets show that the performance of the proposed CBP-Net is superior to the state-of-the-art methods.

Ship Targets Detection in Remote Sensing Images Based on Improved Faster-RCNN

An improved algorithm is proposed to solve the problems of inaccurate recognition and low recall of Faster-Regions with Convolutional Neural Network (Faster-RCNN) algorithm for the detection of ship targets in remote sensing images. The algorithm is based on the Faster-RCNN network framework. Aiming at the small size and dense distribution of ship targets in remote sensing images, the feature extraction network is improved to enhance the detection ability of small targets. ResNet50 is used as the basic feature extraction network of the algorithm,and the hole residual block is introduced for multi-layer feature fusion to construct a new feature extraction network,which improves the feature extraction capability of the algorithm. The experimental results show that compared with the Faster-RCNN algorithm, this algorithm can learn more abundant target features in smaller pixel areas, thereby effectively improving the detection accuracy of ship targets.

Parallel Ensemble Deep Learning for Real-Time Remote Sensing Video Multi-Target Detection

Unmanned aerial vehicle (UAV) is one of the main means of information warfare, such as in battlefield cruises, reconnaissance, and military strikes. Rapid detection and accurate recognition of key targets in UAV images are the basis of subsequent military tasks. The UAV image has characteristics of high resolution and small target size, and in practical application, the detection speed is often required to be fast. Existing algorithms are not able to achieve an effective trade-off between detection accuracy and speed. Therefore, this paper proposes a parallel ensemble deep learning framework for unmanned aerial vehicle video multi-target detection, which is a global and local joint detection strategy. It combines a deep learning target detection algorithm with template matching to make full use of image information. It also integrates multi-process and multi-threading mechanisms to speed up processing. Experiments show that the system has high detection accuracy for targets with focal lengths varying from one to ten times. At the same time, the real-time and stable display of detection results is realized by aiming at the moving UAV video image.