Robotic grasping method of bolster spring based on image-based visual servoing with YOLOv3 object detection algorithm

In this paper, to address the problem of automatic positioning and grasping of bolster spring with complex geometric features and cluttered background, a novel image-based visual servoing (IBVS) control method based on the corner points features of YOLOv3 object detection bounding box is proposed and applied to the robotic grasping system of bolster spring. The YOLOv3 object detection model is used to detect and position the bolster spring and then based on the corner points features of the bolster spring bounding box, the IBVS controller is designed to drive the end effector of the robot to the desired pose. This method adopts the approach-align-grasp control strategy to achieve the grasping of the bolster spring, which is very robust to the calibration parameter errors of the camera and the robot model. With the help of Robotics and Machine Vision Toolboxes in Matlab, IBVS simulation analysis based on feature points is carried out. The results show that it is reasonable to use the corner points of the object detection bounding box as image features under the initial pose where the image plane is parallel to the object plane. The positioning and grasping experiments are conducted on the robotic grasping platform of bolster spring. The results show that this method is effective for automatic positioning and grasping of bolster spring with complex geometric features and cluttered background, and it has strong robustness to the change of illumination.

Genetic Feature Fusion for Object Skeleton Detection

Object skeleton detection requires the convolutional neural networks to recognize objects and their parts in the cluttered background, overcome the image definition degradation brought by the pooling layers, and predict the location of skeleton pixels in different scale granularity. Most existing object skeleton detection methods take great efforts into the designing of side-output networks for multiscale feature fusion. Despite the great progress achieved by them, there are still many problems that hinder the development of object skeleton detection, such as the manually designed network is labor-intensive and the network initialization depends on models pretrained on large-scale datasets. To alleviate these issues, we propose a genetic NAS method to automatically search on a newly designed architecture search space for adaptive multiscale feature fusion. Furthermore, we introduce a symmetric encoder-decoder search space based on reversing the VGG network, in which the decoder can reuse the ImageNet pretrained model of VGG. The searched networks improve the performance of the state-of-the-art methods on commonly used skeleton detection benchmarks, which proves the efficacy of our method.

Multiple Detection and Tracking of Multi Class Vehicles using Locality Sensitive Histogram.

Multiple object detection and tracking in a cluttered background is most important in vision-based applications. In this paper, the goal is to develop a classifier that detects and tracks multiple objects thereby ensuring robustness and accuracy. Locality Sensitive Histogram feature extraction is used, which adds contributions from all the pixels in an image. These features extracted are trained using decision tree classifier which performs with an accuracy of 97%. Experimental results demonstrate the objects tracked and detected under different scale and pose variations. Evaluation and comparison of the proposed method with various other techniques is performed using performance parameters. Results depict that the proposed technique outperforms with increased accuracy and is the top performer

Efficient Hybrid Supervision for Instance Segmentation in Aerial Images

Instance segmentation in aerial images is of great significance for remote sensing applications, and it is inherently more challenging because of cluttered background, extremely dense and small objects, and objects with arbitrary orientations. Besides, current mainstream CNN-based methods often suffer from the trade-off between labeling cost and performance. To address these problems, we present a pipeline of hybrid supervision. In the pipeline, we design an ancillary segmentation model with the bounding box attention module and bounding box filter module. It is able to generate accurate pseudo pixel-wise labels from real-world aerial images for training any instance segmentation models. Specifically, bounding box attention module can effectively suppress the noise in cluttered background and improve the capability of segmenting small objects. Bounding box filter module works as a filter which removes the false positives caused by cluttered background and densely distributed objects. Our ancillary segmentation model can locate object pixel-wisely instead of relying on horizontal bounding box prediction, which has better adaptability to arbitrary oriented objects. Furthermore, oriented bounding box labels are utilized for handling arbitrary oriented objects. Experiments on iSAID dataset show that the proposed method can achieve comparable performance (32.1 AP) to fully supervised methods (33.9 AP), which is obviously higher than weakly supervised setting (26.5 AP), when using only 10% pixel-wise labels.

ℱ3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images

Object detection in remote sensing (RS) images is a challenging task due to the difficulties of small size, varied appearance, and complex background. Although a lot of methods have been developed to address this problem, many of them cannot fully exploit multilevel context information or handle cluttered background in RS images either. To this end, in this paper, we propose a feature fusion and filtration network (F3-Net) to improve object detection in RS images, which has higher capacity of combining the context information at multiple scales while suppressing the interference from the background. Specifically, F3-Net leverages a feature adaptation block with a residual structure to adjust the backbone network in an end-to-end manner, better considering the characteristics of RS images. Afterward, the network learns the context information of the object at multiple scales by hierarchically fusing the feature maps from different layers. In order to suppress the interference from cluttered background, the fused feature is then projected into a low-dimensional subspace by an additional feature filtration module. As a result, more relevant and accurate context information is extracted for further detection. Extensive experiments on DOTA, NWPU VHR-10, and UCAS AOD datasets demonstrate that the proposed detector achieves very promising detection performance.

Microsaccades during high speed continuous visual search

Here, we provide an analysis of the microsaccades that occurred during continuous visual search and targeting of small faces that we pasted either into cluttered background photos or into a simple gray background. Subjects continuously used their eyes to target singular 3-degree upright or inverted faces in changing scenes. As soon as the participant’s gaze reached the target face, a new face was displayed in a different and random location. Regardless of the experimental context (e.g. background scene, no background scene), or target eccentricity (from 4 to 20 degrees of visual angle), we found that the microsaccade rate dropped to near zero levels within only 12 milliseconds after stimulus onset. There were almost never any microsaccades after stimulus onset and before the first saccade to the face. One subject completed 118 consecutive trials without a single microsaccade. However, in about 20% of the trials, there was a single microsaccade that occurred almost immediately after the preceding saccade’s offset. These microsaccades were task oriented because their facial landmark targeting distributions matched those of saccades within both the upright and inverted face conditions. Our findings show that a single feedforward pass through the visual hierarchy for each stimulus is likely all that is needed to effectuate prolonged continuous visual search. In addition, we provide evidence that microsaccades can serve perceptual functions like correcting saccades or effectuating task-oriented goals during continuous visual search.