Significance Attention, observed in a wide variety of animals from insects to humans, involves selectively attending behaviorally relevant stimuli while filtering out other stimuli. We designed paradigm that allowed us record brain activity tethered, walking bees selecting virtual visual objects. found stimulus-specific increased when the controlled position objects, and decreased were not control. When presented with competing optic lobes preceded behavioral choices; this suggests tiny...

Automated robotic platforms are an important part of precision agriculture solutions for sustainable food production. Agri-robots require robust and accurate guidance systems in order to navigate between crops from their base station. Onboard sensors such as machine vision cameras offer a flexible alternative more expensive structured environments scanning lidar or RTK-GNSS. The main challenges visual crop row the dramatic differences appearance farms throughout season variations spacing...

We describe a guidance system for achieving automatic landing of fixed‐wing aircraft in unstructured outdoor terrain, using onboard video cameras. The uses optic flow information sensing and controlling the height above ground, on horizon profile, also acquired by vision stabilizing roll pitch, additionally, if required, control stabilization yaw flight direction. At low heights, when is unreliable, stereo used to guide descent close touchdown. While rate gyro augment attitude one designs,...

This paper describes a technique for estimating the attitude of UAV by monitoring visual horizon. An algorithm is developed that makes best use color and intensity information in an image to determine position orientation horizon, infer aircraft's attitude. The accurate, reliable, fully capable real-time operation. Furthermore, it can be incorporated into any existing vision system, irrespective way which environment imaged (e.g. through lenses or mirrors).

Abstract Background Treating prostate cancer with radiation therapy in patients a history of prior pelvic may be limited by rectal dose constraints and the risk toxicity. Rectal spacers have been shown to improve dosimetry treatment cancer. The study aimed evaluate safety outcomes hydrogel spacer placement, specifically SpaceOAR, between rectum who had previously undergone therapy. Methods In this retrospective case series, we analysed medical records eight sequential undergoing...

In the years to come, large power grid operators will operate and maintain an ever-increasing asset base. New innovative solutions are needed increase quality efficiency of management avoid corresponding growth in resources cost. To this end, autonomous unmanned aerial vehicles (UAVs) provide a range possibilities. Here, we present novel prototype solution for remotely operated inspection missions with resident drones on electrical substations, comprising: (1) drone sense robustness harsh...

This study describes a novel, vision-based system for guidance of UAVs. The uses two cameras, each associated with specially-shaped reflective surface, to obtain stereo information on the height above ground and distances potential obstacles. camera-mirror has advantage that it remaps world onto cylindrical co-ordinate simplifies speeds up range computations, defines collision-free cylinder through which aircraft can pass without encountering result is computationally efficient approach...

This study describes the design and implementation of several bioinspired algorithms for providing guidance to an ultra-lightweight micro-aerial vehicle (MAV) using a 2.6 g omnidirectional vision sensor. Using this visual system we demonstrate autonomous speed control, centring, heading stabilisation on board 30 MAV flying in corridor-like environment. In addition computation wide-field optic flow, comparatively high-resolution imagery provided by sensor also offers potential image-based...

When using virtual-reality paradigms to study animal behaviour, careful attention must be paid how the animal's actions are detected. This is particularly relevant in closed-loop experiments where interacts with a stimulus. Many different sensor types have been used measure aspects of and although some sensors may more accurate than others, few studies examined whether, how, such differences affect an behaviour experiment. To investigate this issue, we conducted tethered honeybees walking on...

For robotic technology to be adopted within the agricultural domain, there is a need for low-cost systems that can flexibly deployed across wide variety of crop types, environmental conditions, and planting methods, without extensive re-engineering. Here we present an approach predicting steering angles autonomous, row-following, agri-robot using only RGB image input. Our employs deep convolutional neural network (DCNN) end-to-end learning strategy. We pre-train our existing open datasets...

This study describes a novel, vision-based system for guidance of UAVs. The uses two coaxially aligned cameras, each associated with specially-shaped reflective surface, to obtain stereo information on the height above ground and distances potential obstacles. camera-mirror has advantage that it remaps world onto cylindrical co-ordinate simplifies speeds up range computations, defines collision-free cylinder through which aircraft can pass without encountering We describe an approach, using...

This study describes a novel method for estimating the strength and direction of local wind field from mobile airborne platform. An iterative optimisation is derived that allows properties to be determined successive measurements heading ground track aircraft only. We have previously described methods these parameters using single vision system. approach therefore constitutes purely visual field. present results simulated real-world flight tests demonstrate accuracy robustness proposed its...

When developing a predictive tool for human performance one needs to have clear metrics evaluate the model's performance. In area of Visual Attention Modeling (VAM) typically compares eye-tracking data collected on group observers predictions made by model. To these models uses signal detection (Receiver Operating Characteristic (ROC)) that measures power system comparing an image eye tracking data. These ROC curves take into account hit and false alarm rates averaging over set test images...

In this paper we present how to exploit a combination of an underwater range-gated imaging system and unsharp filtering method effectively suppress the effects forward- backward-scatter. Typically, back-scattered light causes contrast reduction in image, while forward-scattered blurring features. A can majority back-scatter by gating close objects interest, however, such systems do not have inherent way suppressing forward-scatter. The proposed facilitates real-time 3D because it acquisition...

We propose an approach for robot-supervised learning that automates label generation semantic segmentation with Convolutional Neural Networks (CNNs) crop row detection in a field. Using training robot equipped RTK GNSS and RGB camera, we train neural network can later be used pure vision-based navigation. test our on agri-robot strawberry field successfully without any hand-drawn image labels. Our main finding is the resulting output of CNN shows better performance than noisy labels it was...

Hydrogel spacers aim to separate the rectum from prostate during radiation therapy for patients with cancer decrease dose and thus toxicity rectum. The of this study was evaluate distribution hydrogel spacer between prostate, assess rectal wall infiltration immediate complications.

This study describes a novel method for automatically obtaining the attitude of an aircraft from visual horizon. A wide-angle view environment, including horizon, is captured and input images are classified into fuzzy sky ground regions using spectral intensity properties pixels. The classifier updated continuously online reinforcement strategy therefore able to adapt changing appearance ground, without requiring prior training offline. approach described, which reliable, accurate,...