In conclusion, contrary to the GLM approach, TICA provides sharp insight into the dynamically evolving functional brain connectivity mobilized by the whole experimental paradigm. This multivariate model-free technique is able to differentiate activation of task-specific sequentially or concomitantly re- cruited networks and deactivation of task-unrelated or resting state networks. This study shows that the tactile-tactile match- ing discrimination task specifically activated 3 cooperative networks, including a sensorimotor network (motor program tk;4and execution), executive network (on-line elaboration and correction of haptic representation, working memory and sensory-guidance of exploratory movements), and a cerebel- loparietal network (possibly sensory expectations) and deac- tivated the default-mode network and occipital areas and 2 transiently activated audiomotor and audiovisual circuits.
Vitamin C (ascorbic acid), an ‘over the counter’ supplement, has numerous physiological functions and it is found in high concentrations in the brain. The effect of vitamin C on cognitive memory and visuospatial memory was studied using the Novel Objectrecognitiontask (NORT) and the Morris water maze (MWM) respectively. Twenty Swiss white albino (CD1) mice, within the age of 90-120 days, were randomly divided into two groups of ten mice each. Mice in group 1 served as the control and so received normal saline orally while the other group received vitamin C (200 mg/kg) orally for 21 days. All animals had access to feed and water ad libitum. Behavioural testing started on day 21. There was no significant difference in swim latencies between the control and test groups in the MWM though there was a uniform reduction in swim latency in both groups during acquisition and reversal training days. There also was no significant difference in quadrant duration and swim latencies of both groups in the probe trial and the visible platform task. The habituation
and prefrontal cortex (PFC) are strong possibilities. To date, hippocampal deficits have been well-characterised; there does however remain a lack of insight into the nature of prefrontal participation. Here, we used a PFC-supported temporal order memory paradigm to examine if IFN- α treatment induced deficits in performance; additionally, we used an objectrecognitiontask to assess the integrity of the perirhinal cortex (PRH). Finally, the utility of exercise as an ameliorative strategy to recover temporal order deficits in rats was also explored.
In the present study, we have focused upon exploring the potential of Aqueous liquorice extract (ALE- 400mg/kg) and Glabridin rich extracts at two dose levels (5mg/kg & 10mg/kg) in reversing the memory deficits. Amnesia was induced in mice and rats by intraperitoneal injection of Scopolamine / Diazepam in Objectrecognitiontask (ORT) and Elevated plus maze (EPM) models. Piracetam, the established nootropic agent was used as a standard in the present study.
memory of LPS-treated animals, supporting a role for NMDA receptors in this effect of spermine [72-74]. Interestingly, these results are in agreement with the study by Kranjac and colleagues , who have shown that partial NMDA receptor agonist D-cycloserine rescues memory consolidation following systemic bacterial endo- toxin exposure. Furthermore, Velloso and colleagues  have found that post-training intrastriatal administration of spermine reverses the recognition memory deficits in the novel objectrecognitiontask induced by quinolinic acid, a model of Huntington’ s disease. It is worth noting that spermine improved memory per se in the novel objectrecognitiontask. To our knowledge, this is the first study showing that spermine improves memory. The finding that ifenprodil prevents the promnesic effect of spermine tempts us to propose that it may involve the same molecu- lar targets proposed for spermidine [33,34,36,38]. How- ever, further studies are necessary to clarify this point.
Orange (Citrus sinensis) is well known for its nutritional and medicinal properties throughout the world. This study was undertaken to test the effect of orange juice on the memory of mice. A total of 228 mice divided in 38 groups were employed in this study. Orange juice (2.5%, 5%, 7.5%, 10% v/v) was administered orally to mice with the help of an oral feeding needle. Morris water maze, elevated plus maze and objectrecognitiontask served as exteroceptive behavioral models, whereas scopolamine -induced amnesia and alprazolam -induced amnesia served as interoceptive memory models. Brain AchE activity was measured using Ellman’s method. Orange juice significantly (p< 0.01) reduced transfer latency of mice in Elevated plus maze model. In another experimental model, orange juice significantly (p<0.01) increased discrimination index in objectrecognitiontask, revealing its memory enhancing potential. Furthermore, orange juice in different concentrations enhanced the time spent by mice in target quadrant, when tested using Morris water maze. In the present study, Brain AChE activity was significantly reduced by orange juice, thereby suggesting pro-cholinergic mechanism for orange juice. These findings, when taken together indicate that orange juice possesses promising memory enhancing potential.
spatial learning and memory in the Morris water maze and displaced objectrecognitiontask in mice, but did affect novel objectrecognition. This is consistent with pre- vious findings that systemic administration of selective α7nAchR agonists reverse working memory impairments caused by NMDAR blockade in several behavioral tasks including the 16-arm radial maze, Y-maze, Morris water maze and linear maze, and novel objectrecognition test [28-31]. There are also other examples of functional inter- action between the α7nAchR and NMDAR. Cholinergic in- nervation of the hippocampus modulates activity-dependent synaptic plasticity, such as long-term potentiation (LTP) and other processes that contribute to learning and mem- ory . Nicotine was found to enhance LTP of EPSPs in the dentate gyrus and to convert weak stimuli-evoked short-term potentiation into LTP in the CA1. The selective α7 nAchR agonists choline and 2,4-dimethoxybenzlidine anabaseine have also been found to mimic the facilitative action of nicotine in potentiating LTP [33-35], although the mechanisms underlying the effects of α7nAchR on NMDAR-mediated function remain unclear.
The developmental psychology brings to the table several contributions to this discussion. For example, there are some evidences supporting the hypothesis that children with ADHD can benefit with working memory training, which leads to improvement in terms of focus of attention (Holmes et al., 2010). These findings suggest that environment and experience are very important to the development of perception and attention. Regarding human perception, the ability to recognize a face among several different faces is one of the most intriguing ab- ilities of humans. Since birth, we have encountered thousands of faces everyday that vary in terms of identity, gender, age, and race. However the question whether face recognition has its own biological foundation or is learned through experience remains unanswered. Face recognition involves the processing and interpretation of a set of visual stimuli, which depend on several factors such as the gender, age, and race of the face.
The problem of objectrecognition has been studied exhaustively in the literature. Some of the earliest work in objectrecognition was done by Marr. Marr’s primal sketch involves extraction of local features which are combined in an hierarchical way into more complex descriptions. He believed that the goal of vision is to reconstruct the 3D scene. Marr and Nishihara  proposed that objects should be represented using an object-centered approach using 3D parts or volumes. These representations are completely view-point in- variant. This led to the theory of “Recognition by Components” (also called the structural description model)  by Biederman who built on Marr’s work of object-centered repre- sentations. He proposed that objects were represented by simple primitives and described syntactic ways to describe the relationships between them.
The efficiency and quality of a feature descriptor are critical to the user experience of many computer vision applications. However, the existing descriptors are either too computationally expensive to achieve real-time performance, or not sufficiently distinctive to identify correct matches from a large database with various transformations. In this paper, we propose a highly efficient and distinctive binary descriptor, called local difference binary (LDB). LDB directly computes a binary string for an image patch using simple intensity and gradient difference tests on pair wise grid cells within the patch. A multiple-gridding strategy and a salient bit-selection method are applied to capture the distinct patterns of the patch at different spatial granularities. Experimental results demonstrate that compared to the existing state-of-the-art binary descriptors, primarily designed for speed, LDB has similar construction efficiency, while achieving a greater accuracy and faster speed for mobile objectrecognition and tracking tasks.
This project is about evaluating hand tools for the purpose of an object handling task. The purpose for this project is to evaluate hand tools that are usually use for handling hot objects. There were two types of hand tool evaluated, the tong (old design) and the proposed design. Those tools were evaluated in terms of ergonomics, usability, and aesthetics. A questionnaire was developed consisting of question related to ergonomics, usability and aesthetics. 50 respondents consisting of representative users of the tools were sampled. The result indicated that the tong (old design) was rated less visually appealing than the new proposed design. In addition, the new design was perceived to be better in terms of ergonomics and in terms of usability as well.
Objectrecognition is one of the fundamental tasks of computer vision. Recent advances in the field enable reliable 2D detections from a single cluttered image. However, many challenges still remain. Object detection needs timely response for real world applications. Moreover, we are genuinely interested in estimating the 3D pose and shape of an object or human for the sake of robotic manipulation and human-robot interaction. In this thesis, a suite of solutions to these challenges is presented. First, Active Deformable Part Models (ADPM) is proposed for fast part-based object detection. ADPM dramatically accelerates the detection by dynamically scheduling the part evaluations and efficiently pruning the image locations. Second, we unleash the power of marrying discriminative 2D parts with an explicit 3D geometric representation. Several methods of such scheme are proposed for recovering rich 3D information of both rigid and non-rigid objects from monocular RGB images. (1) The accurate 3D pose of an object instance is recovered from cluttered images using only the CAD model. (2) A global optimal solution for simultaneous 2D part localization, 3D pose and shape estimation is obtained by optimizing a unified convex objective function. Both appearance and
Recent prominent work on Visual ObjectRecognition can be seen in the project of the self-driving car made by Google-Waymo. TensorFlow is one of the promising API which supports ObjectRecognition, a better way computationally for Deep Learning and Machine Learning. Intel is currently working on the project of the autonomous vehicles and for such intelligent system which uses the concept of objectrecognition, their team is working on their own GPU and GPU frameworks. Inspired by the work of Viola and Jones, Mehul K Dabhi and Bhavna K Pancholi improved and tested the face detector implemented Haar features for the feature extraction. Highly motivated by interesting work of the authors Ronald J. Brachman and Thomas G. Dietterich for the better understanding of concepts and contents. The main goal of this paper is to illuminate the theory and concepts of how generic objectrecognition works in the machine-learning establishment.
DOI: 10.4236/jsip.2018.92006 95 Journal of Signal and Information Processing LeCun et al . have applied CNNs to the task of objectrecognition in  and they have shown that CNNs performed the best compared to other object recog- nition techniques. Hinton et al .  have applied deep CNNs to the task of image classification on the Imagenet database and they have achieved the best perfor- mance on the ImageNet Large Scale Visual Recognition Challenge (ILSVRC). The work of Hinton et al . thus ushered a flurry of research activities in the field of deep learning using deep CNNs for image recognition and classification tasks -. Gu et al . have presented a good survey on recent advances in CNNs in . Since we have considered the benchmark Caltech101 database in this work, we discuss here a review of the major research efforts conducted by researchers using this database. Simonyan et al . have applied very deep convolutional net- works for large scale image recognition purposes and they have made their best performing deep learning networks i.e . the VGG-16 and VGG-19 public for other researchers . They have also shown in their work that their pre-trained models generalize well to other datasets such as Caltech-101 and Caltech-256. For generalization purposes, they have considered the penultimate layer for fea- ture extraction and have used an SVM based external classifier to achieve a mean class recall of 92.7 ± 0.5 on the Caltech-101 database . Zeiler et al .  have reported a classification accuracy of 86.5% on the Caltech-101 database in their work on visualization and understanding convolutional networks. They have presented strong useful feature visualizations and activations in their work. They have also discussed the strong generalization of features from one system trained on the ImageNet database to other databases such Caltech-101. Our current work builds on these concepts by considering multiple deeper layers of pre-trained CNN models for feature extraction, their subsequent dimensionality reduction and by selecting the best layer features based on performance of our Classifier Ensemble. He et al .  have considered the problem of visual recog- nition by applying Spatial Pyramid Pooling to Deep Convolutional Networks. They have applied their technique to the Caltech-101 database and have achieved a classification accuracy of 91.44%. Convolutional nets have been ap- plied by Chatfield et al . in  to Caltech-101 database and a recognition accu- racy of 88.4% has been achieved by their system. Object categorization through Group-Sensitive Multiple Kernel Learning has been considered by Yang et al . in . They have reported a recognition performance of 84.3% on the Caltech-101 database.
The classification of deformable objects is an important aspect of computer vision and image processing. Features are the most important aspect of this classification. With- out robust features the classification cannot be successful. Deformable objects complicate the extraction of features by creating a range of values over which the features will exist. It is very likely that objects, particularly natural objects, will not maintain the same appearance over the course of that object’s lifetime. This is especially true of objects that undergo locomotion. The kinematics of the body will cause multiple local deforma- tions across the periphery of the object. These must be expected by the system in order to correctly classify these creatures. The first step to recognizing these deformations is utilizing advanced techniques to represent the salient features of each object.
Experiment 4a showed that under low perceptual load, distracter objects produced repetition priming even if presented in mirror- reflected views between the prime and probe displays (c.f., Stank- iewicz et al., 1998) [Table 4]. However, invariance across mirror reflection may not indicate full viewpoint invariance. Some view- selective neurons are tuned to both mirror reflection views of an object (Logothetis, Pauls, & Poggio, 1995) and observers often show invariance across mirror reflection (e.g., Biederman & Coo- per, 1991; Fiser & Biederman, 2001; Seamon et al., 1997). In contrast, depth rotations typically show costs in recognition per- formance (e.g., Thoma & Davidoff, 2006). In Experiment 4b, we provided a stronger test for our claim that distracter representations in conditions of low perceptual load include structural descrip- tions. We presented photo-realistic images of 3D distracter objects from different viewpoints in 3D space (avoiding a mirror- reflection change in view). A view-invariant distracter recognition in this experiment would demonstrate that observers formed struc- tural representations of these distracter objects.
Figure 3.19 shows an example of the non-detection process that is observed most frequently. Panel (a) shows the test image, (b) the model of the corresponding object. The large blue part is flexible, its structure changes completely between both shots, which makes the task difficult for any detection system based on pose. On the other hand, the white handle and the colored bells are rigid parts which should be matched reliably between views. However, their smooth surface triggers only very few features detections with poor localization. Most features detected in this picture are generated by the grainy background of concrete, which create random candidate matches with all models as can be seen in panel (c). The textureless and specular object surface make the orientation and scale information in the SIFT descriptor unreliable. As a result, the candidate matches are spread in the whole Hough space, and the Hough transform stage identifies only one cluster with enough matches to continue to the next detection stage. This cluster is in fact incorrect, and rejected by the PROSAC stage. One might object that the values of the thresholds T k
object present in an image or a video sequence, with the help of some recognition technique or methods. Objectrecognition is one of the techniques of digital image processing where we can process any image by applying some of the operation. The modern world is enclosed with gigantic masses of digital visual information. Increase in the images has urged for the development of robust and efficient objectrecognition techniques. Most work reported in the literature focuses on competent techniques for objectrecognition and its applications.
Machine vision is one of the applications of computer vision or image processing to industry and manufacturing scenario aiming at automation of processes like object classification, sorting, object counting and monitoring. One of the most common applications of machine vision is the inspection of packages, cargos, manufactured goods such as semiconductors chips, foods and pharmaceuticals. This inspection and classification is based on the fact that each of the object that is monitored has distinguishing features from the others, in terms of shape descriptors like extent, aspect ratio or colour. This uniqueness is what makes them discernible from each other. Thus the proposed ARM controller based objectrecognition is based on fundamentals of machine vision wherein by image processing, counting and identification of all the shapes and their colours in an image is possible. The system is implemented using Raspberry Pi 2 model B and the algorithm is mainly based on the techniques of image segmentation and blob analysis. This identifies objects in the image that are clearly discernible from the background.