Machine Learning Approach to Detect Drowsiness on Behavioral Parameters

Object detection has received a lot of research attention in recent years because of its tight association with video analysis and picture interpretation. Face detection, vehicle detection, pedestrian counting, web photos, security systems, and self-driving automobiles are all examples of object detection. With little conscious thought, the human visual system can accomplish complicated tasks such as distinguishing multiple objects and detecting impediments. Thanks to the availability of large amounts of data, faster GPUs, and improved algorithms, we can now quickly train computers to detect and classify many elements inside a picture with high accuracy. Our project is focused on building a single-access platform for various object detection tasks. A user-interface where the user is asked for the relevant inputs and an output based on this is generated automatically by the system. Also, accuracy and precision measures are also displayed so that the user is wary of their liability extent on the generated results.

Seeing the World From Others’ Perspective: 14-Month-Olds Show Altercentric Modulation Effects by Others’ Beliefs

Humans have a propensity to readily adopt others’ perspective, which often influences their behavior even when it seemingly should not. This altercentric influence has been widely studied in adults, yet we lack an understanding of its ontogenetic origins. The current studies investigated whether 14-month-olds’ search in a box for potential objects is modulated by another person’s belief about the box’s content. We varied the person’s potential belief such that in her presence/absence an object was removed, added, or exchanged for another, leading to her true/false belief about the object’s presence (Experiment 1, n = 96); or transformed into another object, leading to her true/false belief about the object’s identity (i.e., the objects represented under a specific aspect, Experiment 2, n = 32). Infants searched longer if the other person believed that an object remained in the box, showing an altercentric influence early in development. These results suggest that infants spontaneously represent others’ beliefs involving multiple objects and raise the possibility that infants can appreciate that others encode the world under a unique aspect.

Broad attention uncovers benefits of stimulus uniformity in visual crowding

Crowding is the interference by surrounding objects (flankers) with target perception. Low target-flanker similarity usually yields weaker crowding than high similarity (‘similarity rule’) with less interference, e.g., by opposite- than same-contrast polarity flankers. The advantage of low target-flanker similarity has typically been shown with attentional selection of a single target object. Here, we investigated the validity of the similarity rule when broadening attention to multiple objects. In three experiments, we measured identification for crowded letters (Experiment 1), tumbling Ts (Experiment 2), and tilted lines (Experiment 3). Stimuli consisted of three items that were uniform or alternating in contrast polarity and were briefly presented at ten degrees eccentricity. Observers reported all items (full report) or only the left, central, or right item (single-item report). In Experiments 1 and 2, consistent with the similarity rule, single central item performance was superior with opposite- compared to same-contrast polarity flankers. With full report, the similarity rule was inverted: performance was better for uniform compared to alternating stimuli. In Experiment 3, contrast polarity did not affect performance. We demonstrated a reversal of the similarity rule under broadened attention, suggesting that stimulus uniformity benefits crowded object recognition when intentionally directing attention towards all stimulus elements. We propose that key properties of crowding have only limited validity as they may require a-priori differentiation of target and context.

Inhibitory Control in Age-Related Hearing Loss

Inhibitory control involves suppression of goal irrelevant information and responses. Emerging evidence suggests alterations in inhibitory control in individuals with age-related hearing loss (ARHL), however, few have specifically studied individuals with mild ARHL. We examined behavioral and event related potential (ERP) differences between 14 older adults with mild ARHL (mean age: 69.43 ± 7.73 years) and 14 age- and education-matched normal hearing (NH, mean age: 66.57 ± 5.70 years) controls on two Go/NoGo tasks: a simpler, basic categorization task (Single Car; SC) and a more difficult, superordinate categorization task (Object Animal; OA). The SC task consisted of exemplars of a single car and dog, and the OA task consisted of exemplars of multiple objects and animals. Participants were required to respond to Go trials (e.g., cars in SC) with a button press, and withhold responses on NoGo trials (e.g., dogs in SC task). Behavioral results revealed that ARHL group had worse accuracy on NoGo trials on the OA task, but not on the SC task. ARHL group had longer N2 latency for NoGo compared to Go trials in the simpler SC Task, but no differences were observed on the OA task between Go and NoGo trials. These findings suggest that more prolonged neural effort in the ARHL group on the SC task NoGo trials may have contributed to their ability to successfully suppress false alarms comparable to the NH group. Overall, these findings provide evidence for behavioral and neural changes in inhibitory control in ARHL.

Resources Underlying Visuo-Spatial Working Memory Enable Veridical Large Numerosity Perception

Humans can quickly approximate how many objects are in a visual image, but no clear consensus has been achieved on the cognitive resources underlying this ability. Previous work has lent support to the notion that mechanisms which explicitly represent the locations of multiple objects in the visual scene within a mental map are critical for both visuo-spatial working memory and enumeration (at least for relatively small numbers of items). Regarding the cognitive underpinnings of large numerosity perception, an issue currently subject to much controversy is why numerosity estimates are often non-veridical (i.e., susceptible to biases from non-numerical quantities). Such biases have been found to be particularly pronounced in individuals with developmental dyscalculia (DD), a learning disability affecting the acquisition of arithmetic skills. Motivated by findings showing that DD individuals are also often impaired in visuo-spatial working memory, we hypothesized that resources supporting this type of working memory, which allow for the simultaneous identification of multiple objects, might also be critical for precise and unbiased perception of larger numerosities. We therefore tested whether loading working memory of healthy adult participants during discrimination of large numerosities would lead to increased interference from non-numerical quantities. Participants performed a numerosity discrimination task on multi-item arrays in which numerical and non-numerical stimulus dimensions varied congruently or incongruently relative to each other, either in isolation or in the context of a concurrent visuo-spatial or verbal working memory task. During performance of the visuo-spatial, but not verbal, working memory task, precision in numerosity discrimination decreased, participants’ choices became strongly biased by item size, and the strength of this bias correlated with measures of arithmetical skills. Moreover, the interference between numerosity and working memory tasks was bidirectional, with number discrimination impacting visuo-spatial (but not verbal) performance. Overall, these results suggest that representing visual numerosity in a way that is unbiased by non-numerical quantities relies on processes which explicitly segregate/identify the locations of multiple objects that are shared with visuo-spatial (but not verbal) working memory. This shared resource may potentially be impaired in DD, explaining the observed co-occurrence of working memory and numerosity discrimination deficits in this clinical population.

Comparison of deep learning models in terms of multiple object detection on satellite images

The images obtained by remote sensing contain important data about ground surface. It is an important issue to detect objects on the ground surface with these images. Deep learning models are known to give better results in studies on object detection. However, the superiority of the deep learning models over each other is unknown. For this reason, it should be clarified which model is superior in terms of object detection and which model should be used in studies. In this study, it was aimed to reveal the superiorities of deep learning models by comparing their performance in detecting multiple objects. By using 11 deep learning models that are frequently encountered in the literature, the application of detecting objects of 14 classes in the DOTA dataset were made. 49,053 objects in 888 images were used for training by using AlexNet, Vgg16, Vgg19, GoogleNet, SequezeeNet, Resnet18, Resnet50, Resnet101, Inceptionresnetv2, inceptionv3, DenseNet201 models. After the training, 13,772 objects consisting of 14 classes in 277 images were used for testing with RCNN, which is one of the object detection methods. The performance of each algorithm in 14 classes has been demonstrated by using Average Precision (AP) and Mean Average Precision (mAP) to measure the performance of the models from their metrics. In a particular class of each deep learning model, difference in performance was observed The model with the highest performance varies in each class. In the application, the most successful average mAP value of 14 classes was Vgg16 with 24.64, while the lowest was InceptionResnetV2 with 11.78. In this article, the success of deep learning models in detecting multiple objects has been demonstrated practically and it is thought to be an important resource for researchers who will study on this subject.

Planning for grasping cluttered objects based on obstruction degree

Grasping objects in clutter is more difficult than grasping a separated single object. An important issue is that unsafe grasps may occur, in case, one object sits or leans on another, which could cause the collapse of objects. In addition, reachability of each object surrounded by other obstacles also has to be considered. So the order of multiple objects for grasping and the grasp configuration of each object must be planned simultaneously. This article combines grasp order and grasp configuration planning to perform fast and safe multiobject grasping in cluttered scenes. First, a comprehensive grasp configuration database is built to provide enough feasible grasp configurations for the objects. Then, we propose an obstruction degree to estimate the likelihood of reachability of each grasp configuration as well as each object. This measurement also implicitly infers object interactions. Finally, grasp order and grasp configurations are planned together to deal with the constraints caused by reachability and object interaction. Simulations and experiments in a series of cluttered scenes demonstrate that our method can grasp objects efficiently and can greatly reduce unsafe grasps.

Online Multiple Object Tracking Using a Novel Discriminative Module for Autonomous Driving

Multi object tracking (MOT) is a key research technology in the environment sensing system of automatic driving, which is very important to driving safety. Online multi object tracking needs to accurately extend the trajectory of multiple objects without using future frame information, so it will face greater challenges. Most of the existing online MOT methods are anchor-based detectors, which have many misdetections and missed detection problems, and have a poor effect on the trajectory extension of adjacent object objects when they are occluded and overlapped. In this paper, we propose a discrimination learning online tracker that can effectively solve the occlusion problem based on an anchor-free detector. This method uses the different weight characteristics of the object when the occlusion occurs and realizes the extension of the competition trajectory through the discrimination module to prevent the ID-switch problem. In the experimental part, we compared the algorithm with other trackers on two public benchmark datasets, MOT16 and MOT17, and proved that our algorithm has achieved state-of-the-art performance, and conducted a qualitative analysis on the convincing autonomous driving dataset KITTI.