Representation and learning of visual information for pose recognition

Recovering position from sensor information is an important problem in mobile robotics, known as localisation. Localisation requires a map or some other description of the environment to provide the robot with a context to interpret sensor data. The mobile robot system under discussion is using an artificial neural representation of position. Building a geometrical map of the environment with a single camera and artificial neural networks is difficult. Instead it would be simpler to learn position as a function of the visual input. Usually when learning images, an intermediate representation is employed. An appropriate starting point for biologically plausible image representation is the complex cells of the visual cortex, which have invariance properties that appear useful for localisation. The effectiveness for localisation of two different complex cell models are evaluated. Finally the ability of a simple neural network with single shot learning to recognise these representations and localise a robot is examined.

Improving the use of visual interactive simulation as a knowledge elicitation tool

Knowledge elicitation is a well-known bottleneck in the production of knowledge-based systems (KBS). Past research has shown that visual interactive simulation (VIS) could effectively be used to elicit episodic knowledge that is appropriate for machine learning purposes, with a view to building a KBS. Nonetheless, the VIS-based elicitation process still has much room for improvement. Based in the Ford Dagenham Engine Assembly Plant, a research project is being undertaken to investigate the individual/joint effects of visual display level and mode of problem case generation on the elicitation process. This paper looks at the methodology employed and some issues that have been encountered to date. Copyright © 2007 Inderscience Enterprises Ltd.

Knowledge management and the limits of knowledge codification

Purpose - The idea that knowledge needs to be codified is central to many claims that knowledge can be managed. However, there appear to be no empirical studies in the knowledge management context that examine the process of knowledge codification. This paper therefore seeks to explore codification as a knowledge management process. Design/methodology/approach - The paper draws on findings from research conducted around a knowledge management project in a section of the UK Post Office, using a methodology of participant-observation. Data were collected through observations of project meetings, correspondence between project participants, and individual interviews. Findings - The principal findings about the nature of knowledge codification are first, that the process of knowledge codification also involves the process of defining the codes needed to codify knowledge, and second, that people who participate in the construction of these codes are able to interpret and use the codes more similarly. From this it can be seen that the ability of people to decodify codes similarly places restrictions on the transferability of knowledge between them. Research limitations/implications - The paper therefore argues that a new conceptual approach is needed for the role of knowledge codification in knowledge management that emphasizes the importance of knowledge decodification. Such an approach would start with one's ability to decodify rather than codify knowledge as a prerequisite for knowledge management. Originality/value - The paper provides a conceptual basis for explaining limitations to the management and transferability of knowledge. © Emerald Group Publishing Limited.

On the existence of visual technical patterns in the UK stock market

The aim in this paper is to replicate and extend the analysis of visual technical patterns by Lo et al. (2000) using data on the UK market. A non-parametric smoother is used to model a nonlinear trend in stock price series. Technical patterns, such as the 'head-and-shoulders' pattern, that are characterised by a sequence of turning points are identified in the smoothed data. Statistical tests are used to determine whether returns conditioned on the technical patterns are different from random returns and, in an extension to the analysis of Lo et al. (2000), whether they can outperform a market benchmark return. For the stocks in the FTSE 100 and FTSE 250 indices over the period 1986 to 2001, we find that technical patterns occur with different frequencies to each other and in different relativities to the frequencies found in the US market. Our extended statistical testing indicates that UK stock returns are less influenced by technical patterns than was the case for US stock returns.

Distinct contrast response functions in striate and extra-striate regions of visual cortex revealed with magnetoencephalography (MEG)

Objective: To spatially and temporally characterise the cortical contrast response function to pattern onset stimuli in humans. Methods: Magnetoencephalography (MEG) was used to investigate the human cortical contrast response function to pattern onset stimuli with high temporal and spatial resolution. A beamformer source reconstruction approach was used to spatially localise and identify the time courses of activity at various visual cortical loci. Results: Consistent with the findings of previous studies, MEG beamformer analysis revealed two simultaneous generators of the pattern onset evoked response. These generators arose from anatomically discrete locations in striate and extra-striate visual cortex. Furthermore, these loci demonstrated notably distinct contrast response functions, with striate cortex increasing approximately linearly with contrast, whilst extra-striate visual cortex followed a saturating function. Conclusions: The generators that underlie the pattern onset visual evoked response arise from two distinct regions in striate and extra-striate visual cortex. Significance: The spatially, temporally and functionally distinct mechanisms of contrast processing within the visual cortex may account for the disparate results observed across earlier studies and assist in elucidating causal mechanisms of aberrant contrast processing in neurological disorders. © 2005 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

The Effect of nutritional supplementation on subjective and objective measures of visual and retinal function:a random controlled trial

In industrialised countries age-related macular disease (ARMD) is the leading cause of visual loss in older people. Because oxidative stress is purported to be associated with an increased risk of disease development the role of antioxidant supplementation is of interest. Lutein is a carotenoid antioxidant that accumulates within the retina and is thought to filter blue light. Increased levels of lutein have been associated with reduced risk of developing ARMD and improvements in visual and retinal function in eyes with ARMD. The aim of this randomised controlled trial (RCT) was to investigate the effect of a lutein-based nutritional supplement on subjective and objective measures of visual function in healthy eyes and in eyes with age-related maculopathy (ARM) – an early form of ARMD. Supplement withdrawal effects were also investigated. A sample size of 66 healthy older (HO), healthy younger (HY), and ARM eyes were randomly allocated to receive a lutein-based supplement or no treatment for 40 weeks. The supplemented group then stopped supplementation to look at the effects of withdrawal over a further 20 weeks. The primary outcome measure was multifocal electroretinogram (mfERG) N1P1 amplitude. Secondary outcome measures were mfERG N1, P1 and N2 latency, contrast sensitivity (CS), Visual acuity (VA) and macular pigment optical density (MPOD). Sample sizes were sufficient for the RCT to have an 80% power to detect a significant clinical effect at the 5% significance level for all outcome measures when the healthy eye groups were combined, and CS, VA and mfERG in the ARM group. This RCT demonstrates significant improvements in MPOD in HY and HO supplemented eyes. When HY and HO supplemented groups were combined, MPOD improvements were maintained, and mfERG ring 2 P1 latency became shorter. On withdrawal of the supplement mfERG ring 1 N1P1 amplitude reduced in HO eyes. When HO and HY groups were combined, mfERG ring 1 and ring 2 N1P1 amplitudes were reduced. In ARM eyes, ring 3 N2 latency and ring 4 P1 latency became longer. These statistically significant changes may not be clinically significant. The finding that a lutein-based supplement increases MPOD in healthy eyes, but does not increase mfERG amplitudes contrasts with the CARMIS study and contributes to the debate on the use of nutritional supplementation in ARM.

Associative knowledge controls deployment of visual selective attention

According to some models of visual selective attention, objects in a scene activate corresponding neural representations, which compete for perceptual awareness and motor behavior. During a visual search for a target object, top-down control exerted by working memory representations of the target's defining properties resolves competition in favor of the target. These models, however, ignore the existence of associative links among object representations. Here we show that such associations can strongly influence deployment of attention in humans. In the context of visual search, objects associated with the target were both recalled more often and recognized more accurately than unrelated distractors. Notably, both target and associated objects competitively weakened recognition of unrelated distractors and slowed responses to a luminance probe. Moreover, in a speeded search protocol, associated objects rendered search both slower and less accurate. Finally, the first saccades after onset of the stimulus array were more often directed toward associated than control items.

The limits of open innovation:openness and (quasi-)markets in the organization of innovation

Recent studies have stressed the importance of ‘open innovation’ as a means of enhancing innovation performance. The essence of the open innovation model is to take advantage of external as well as internal knowledge sources in developing and commercialising innovation, so avoiding an excessively narrow internal focus in a key area of corporate activity. Although the external aspect of open innovation is often stressed, another key aspect involves maximising the flow of ideas and knowledge from different sources within the firm, for example through knowledge sharing via the use of cross-functional teams. A fully open innovation approach would therefore combine both aspects i.e. cross-functional teams with boundary-spanning knowledge linkages. This suggests that there should be complementarities between the use cross-functional teams with boundary-spanning knowledge linkages i.e. the returns to implementing open innovation in one innovation activity is should be greater if open innovation is already in place in another innovation activity. However, our findings – based on a large sample of UK and German manufacturing plants – do not support this view. Our results suggest that in practice the benefits envisaged in the open innovation model are not generally achievable by the majority of plants, and that instead the adoption of open innovation across the whole innovation process is likely to reduce innovation outputs. Our results provide some guidance on the type of activities where the adoption of a market-based governance structure such as open innovation may be most valuable. This is likely to be in innovation activities where search is deterministic, activities are separable, and where the required level of knowledge sharing is correspondingly moderate – in other words those activities which are more routinized. For this type of activity market-based governance mechanisms (i.e. open innovation) may well be more efficient than hierarchical governance structures. For other innovation activities where outcomes are more uncertain and unpredictable and the risks of knowledge exchange hazards are greater, quasi-market based governance structures such as open innovation are likely to be subject to rapidly diminishing returns in terms of innovation outputs.

Simulation based knowledge elicitation:effect of visual representation and model parameters

Since much knowledge is tacit, eliciting knowledge is a common bottleneck during the development of knowledge-based systems. Visual interactive simulation (VIS) has been proposed as a means for eliciting experts’ decision-making by getting them to interact with a visual simulation of the real system in which they work. In order to explore the effectiveness and efficiency of VIS based knowledge elicitation, an experiment has been carried out with decision-makers in a Ford Motor Company engine assembly plant. The model properties under investigation were the level of visual representation (2-dimensional, 2½-dimensional and 3-dimensional) and the model parameter settings (unadjusted and adjusted to represent more uncommon and extreme situations). The conclusion from the experiment is that using a 2-dimensional representation with adjusted parameter settings provides the better simulation-based means for eliciting knowledge, at least for the case modelled.

Topography of visual evoked magnetic responses to pattern shift, pattern onset and flash stimuli

Different visual stimuli may activate separate channels in the visual system and produce magnetic responses from the human bran which originate from distinct regions of the visual cortex. To test this hypothesis, we have investigated the distribution of visual evoked magnetic responses to three distinct visual stimuli over the occipital region of the scalp with a DC-SQUID second-order gradiometer in an ubshielded environment. Patterned stimuli were presented full field and to the right half field, while a flash stimulus was presented full field only, in five normal subjects. Magnetic responses were recorded from 20 to 42 positions over the occipital scalp. Topographic maps were prepared of the major positive component within the first 150ms to the three stimuli, i.e., the P100m (pattern shift), C11m (pattern onset) and P2m (flash). For the pattern shift stimulus the data suggested the source of the P100m was close to the midline with the current directed towards the medial surface. The data for the pattern onset C11m suggested a source at a similar depth but with the current directed away from the midline towards the lateral surface. The flash P2m appeared to originate closer to the surface of the occipital pole than both the patterned stimuli. Hence the pattern shift (which may represent movement), and the pattern onset C11m (representing contrast and contour) appear to originate in similar areas of brain but to represent different asepcts of cortical processing. By contrast, the flash P2m (representing luminance change) appears to originate in a distinct area of visual cortex closer to the occipital pole.