Maintaining a cognitive map in darkness : the need to fuse boundary knowledge with path integration

Spatial navigation requires the processing of complex, disparate and often ambiguous sensory data. The neurocomputations underpinning this vital ability remain poorly understood. Controversy remains as to whether multimodal sensory information must be combined into a unified representation, consistent with Tolman's "cognitive map", or whether differential activation of independent navigation modules suffice to explain observed navigation behaviour. Here we demonstrate that key neural correlates of spatial navigation in darkness cannot be explained if the path integration system acted independently of boundary (landmark) information. In vivo recordings demonstrate that the rodent head direction (HD) system becomes unstable within three minutes without vision. In contrast, rodents maintain stable place fields and grid fields for over half an hour without vision. Using a simple HD error model, we show analytically that idiothetic path integration (iPI) alone cannot be used to maintain any stable place representation beyond two to three minutes. We then use a measure of place stability based on information theoretic principles to prove that featureless boundaries alone cannot be used to improve localization above chance level. Having shown that neither iPI nor boundaries alone are sufficient, we then address the question of whether their combination is sufficient and - we conjecture - necessary to maintain place stability for prolonged periods without vision. We addressed this question in simulations and robot experiments using a navigation model comprising of a particle filter and boundary map. The model replicates published experimental results on place field and grid field stability without vision, and makes testable predictions including place field splitting and grid field rescaling if the true arena geometry differs from the acquired boundary map. We discuss our findings in light of current theories of animal navigation and neuronal computation, and elaborate on their implications and significance for the design, analysis and interpretation of experiments.

Extracting deck trend to plan descent trajectory for safe landing of UAVs

This paper outlines a feasible scheme to extract deck trend when a rotary-wing unmanned aerial vehicle (RUAV)approaches an oscillating deck. An extended Kalman filter (EKF) is de- veloped to fuse measurements from multiple sensors for effective estimation of the unknown deck heave motion. Also, a recursive Prony Analysis (PA) procedure is proposed to implement online curve-fitting of the estimated heave mo- tion. The proposed PA constructs an appropriate model with parameters identified using the forgetting factor recursive least square (FFRLS)method. The deck trend is then extracted by separating dominant modes. Performance of the proposed procedure is evaluated using real ship motion data, and simulation results justify the suitability of the proposed method into safe landing of RUAVs operating in a maritime environment.

A dual-scale modelling approach for drying hygroscopic porous media

A new dualscale modelling approach is presented for simulating the drying of a wet hygroscopic porous material that couples the porous medium (macroscale) with the underlying pore structure (microscale). The proposed model is applied to the convective drying of wood at low temperatures and is valid in the so-called hygroscopic range, where hygroscopically held liquid water is present in the solid phase and water exits only as vapour in the pores. Coupling between scales is achieved by imposing the macroscopic gradients of moisture content and temperature on the microscopic field using suitably-defined periodic boundary conditions, which allows the macroscopic mass and thermal fluxes to be defined as averages of the microscopic fluxes over the unit cell. This novel formulation accounts for the intricate coupling of heat and mass transfer at the microscopic scale but reduces to a classical homogenisation approach if a linear relationship is assumed between the microscopic gradient and flux. Simulation results for a sample of spruce wood highlight the potential and flexibility of the new dual-scale approach. In particular, for a given unit cell configuration it is not necessary to propose the form of the macroscopic fluxes prior to the simulations because these are determined as a direct result of the dual-scale formulation.

A variable-stepsize Jacobian-free exponential integrator for simulating transport in heterogeneous porous media : application to wood drying

A Jacobian-free variable-stepsize method is developed for the numerical integration of the large, stiff systems of differential equations encountered when simulating transport in heterogeneous porous media. Our method utilises the exponential Rosenbrock-Euler method, which is explicit in nature and requires a matrix-vector product involving the exponential of the Jacobian matrix at each step of the integration process. These products can be approximated using Krylov subspace methods, which permit a large integration stepsize to be utilised without having to precondition the iterations. This means that our method is truly "Jacobian-free" - the Jacobian need never be formed or factored during the simulation. We assess the performance of the new algorithm for simulating the drying of softwood. Numerical experiments conducted for both low and high temperature drying demonstrates that the new approach outperforms (in terms of accuracy and efficiency) existing simulation codes that utilise the backward Euler method via a preconditioned Newton-Krylov strategy.

Performance evaluation of onsite sewage treatment : Evaluation of current research

Efficient management of domestic wastewater is a primary requirement for human well being. Failure to adequately address issues of wastewater collection, treatment and disposal can lead to adverse public health and environmental impacts. The increasing spread of urbanisation has led to the conversion of previously rural land into urban developments and the more intensive development of semi urban areas. However the provision of reticulated sewerage facilities has not kept pace with this expansion in urbanisation. This has resulted in a growing dependency on onsite sewage treatment. Though considered only as a temporary measure in the past, these systems are now considered as the most cost effective option and have become a permanent feature in some urban areas. This report is the first of a series of reports to be produced and is the outcome of a research project initiated by the Brisbane City Council. The primary objective of the research undertaken was to relate the treatment performance of onsite sewage treatment systems with soil conditions at site, with the emphasis being on septic tanks. This report consists of a ‘state of the art’ review of research undertaken in the arena of onsite sewage treatment. The evaluation of research brings together significant work undertaken locally and overseas. It focuses mainly on septic tanks in keeping with the primary objectives of the project. This report has acted as the springboard for the later field investigations and analysis undertaken as part of the project. Septic tanks still continue to be used widely due to their simplicity and low cost. Generally the treatment performance of septic tanks can be highly variable due to numerous factors, but a properly designed, operated and maintained septic tank can produce effluent of satisfactory quality. The reduction of hydraulic surges from washing machines and dishwashers, regular removal of accumulated septage and the elimination of harmful chemicals are some of the practices that can improve system performance considerably. The relative advantages of multi chamber over single chamber septic tanks is an issue that needs to be resolved in view of the conflicting research outcomes. In recent years, aerobic wastewater treatment systems (AWTS) have been gaining in popularity. This can be mainly attributed to the desire to avoid subsurface effluent disposal, which is the main cause of septic tank failure. The use of aerobic processes for treatment of wastewater and the disinfection of effluent prior to disposal is capable of producing effluent of a quality suitable for surface disposal. However the field performance of these has been disappointing. A significant number of these systems do not perform to stipulated standards and quality can be highly variable. This is primarily due to houseowner neglect or ignorance of correct operational and maintenance procedures. The other problems include greater susceptibility to shock loadings and sludge bulking. As identified in literature a number of design features can also contribute to this wide variation in quality. The other treatment processes in common use are the various types of filter systems. These include intermittent and recirculating sand filters. These systems too have their inherent advantages and disadvantages. Furthermore as in the case of aerobic systems, their performance is very much dependent on individual houseowner operation and maintenance practices. In recent years the use of biofilters has attracted research interest and particularly the use of peat. High removal rates of various wastewater pollutants have been reported in research literature. Despite these satisfactory results, leachate from peat has been reported in various studies. This is an issue that needs further investigations and as such biofilters can still be considered to be in the experimental stage. The use of other filter media such as absorbent plastic and bark has also been reported in literature. The safe and hygienic disposal of treated effluent is a matter of concern in the case of onsite sewage treatment. Subsurface disposal is the most common and the only option in the case of septic tank treatment. Soil is an excellent treatment medium if suitable conditions are present. The processes of sorption, filtration and oxidation can remove the various wastewater pollutants. The subsurface characteristics of the disposal area are among the most important parameters governing process performance. Therefore it is important that the soil and topographic conditions are taken into consideration in the design of the soil absorption system. Seepage trenches and beds are the common systems in use. Seepage pits or chambers can be used where subsurface conditions warrant, whilst above grade mounds have been recommended for a variety of difficult site conditions. All these systems have their inherent advantages and disadvantages and the preferable soil absorption system should be selected based on site characteristics. The use of gravel as in-fill for beds and trenches is open to question. It does not contribute to effluent treatment and has been shown to reduce the effective infiltrative surface area. This is due to physical obstruction and the migration of fines entrained in the gravel, into the soil matrix. The surface application of effluent is coming into increasing use with the advent of aerobic treatment systems. This has the advantage that treatment is undertaken on the upper soil horizons, which is chemically and biologically the most effective in effluent renovation. Numerous research studies have demonstrated the feasibility of this practice. However the overriding criteria is the quality of the effluent. It has to be of exceptionally good quality in order to ensure that there are no resulting public health impacts due to aerosol drift. This essentially is the main issue of concern, due to the unreliability of the effluent quality from aerobic systems. Secondly, it has also been found that most householders do not take adequate care in the operation of spray irrigation systems or in the maintenance of the irrigation area. Under these circumstances surface disposal of effluent should be approached with caution and would require appropriate householder education and stringent compliance requirements. However despite all this, the efficiency with which the process is undertaken will ultimately rest with the individual householder and this is where most concern rests. Greywater too should require similar considerations. Surface irrigation of greywater is currently being permitted in a number of local authority jurisdictions in Queensland. Considering the fact that greywater constitutes the largest fraction of the total wastewater generated in a household, it could be considered to be a potential resource. Unfortunately in most circumstances the only pretreatment that is required to be undertaken prior to reuse is the removal of oil and grease. This is an issue of concern as greywater can considered to be a weak to medium sewage as it contains primary pollutants such as BOD material and nutrients and may also include microbial contamination. Therefore its use for surface irrigation can pose a potential health risk. This is further compounded by the fact that most householders are unaware of the potential adverse impacts of indiscriminate greywater reuse. As in the case of blackwater effluent reuse, there have been suggestions that greywater should also be subjected to stringent guidelines. Under these circumstances the surface application of any wastewater requires careful consideration. The other option available for the disposal effluent is the use of evaporation systems. The use of evapotranspiration systems has been covered in this report. Research has shown that these systems are susceptible to a number of factors and in particular to climatic conditions. As such their applicability is location specific. Also the design of systems based solely on evapotranspiration is questionable. In order to ensure more reliability, the systems should be designed to include soil absorption. The successful use of these systems for intermittent usage has been noted in literature. Taking into consideration the issues discussed above, subsurface disposal of effluent is the safest under most conditions. This is provided the facility has been designed to accommodate site conditions. The main problem associated with subsurface disposal is the formation of a clogging mat on the infiltrative surfaces. Due to the formation of the clogging mat, the capacity of the soil to handle effluent is no longer governed by the soil’s hydraulic conductivity as measured by the percolation test, but rather by the infiltration rate through the clogged zone. The characteristics of the clogging mat have been shown to be influenced by various soil and effluent characteristics. Secondly, the mechanisms of clogging mat formation have been found to be influenced by various physical, chemical and biological processes. Biological clogging is the most common process taking place and occurs due to bacterial growth or its by-products reducing the soil pore diameters. Biological clogging is generally associated with anaerobic conditions. The formation of the clogging mat provides significant benefits. It acts as an efficient filter for the removal of microorganisms. Also as the clogging mat increases the hydraulic impedance to flow, unsaturated flow conditions will occur below the mat. This permits greater contact between effluent and soil particles thereby enhancing the purification process. This is particularly important in the case of highly permeable soils. However the adverse impacts of the clogging mat formation cannot be ignored as they can lead to significant reduction in the infiltration rate. This in fact is the most common cause of soil absorption systems failure. As the formation of the clogging mat is inevitable, it is important to ensure that it does not impede effluent infiltration beyond tolerable limits. Various strategies have been investigated to either control clogging mat formation or to remediate its severity. Intermittent dosing of effluent is one such strategy that has attracted considerable attention. Research conclusions with regard to short duration time intervals are contradictory. It has been claimed that the intermittent rest periods would result in the aerobic decomposition of the clogging mat leading to a subsequent increase in the infiltration rate. Contrary to this, it has also been claimed that short duration rest periods are insufficient to completely decompose the clogging mat, and the intermediate by-products that form as a result of aerobic processes would in fact lead to even more severe clogging. It has been further recommended that the rest periods should be much longer and should be in the range of about six months. This entails the provision of a second and alternating seepage bed. The other concepts that have been investigated are the design of the bed to meet the equilibrium infiltration rate that would eventuate after clogging mat formation; improved geometry such as the use of seepage trenches instead of beds; serial instead of parallel effluent distribution and low pressure dosing of effluent. The use of physical measures such as oxidation with hydrogen peroxide and replacement of the infiltration surface have been shown to be only of short-term benefit. Another issue of importance is the degree of pretreatment that should be provided to the effluent prior to subsurface application and the influence exerted by pollutant loadings on the clogging mat formation. Laboratory studies have shown that the total mass loadings of BOD and suspended solids are important factors in the formation of the clogging mat. It has also been found that the nature of the suspended solids is also an important factor. The finer particles from extended aeration systems when compared to those from septic tanks will penetrate deeper into the soil and hence will ultimately cause a more dense clogging mat. However the importance of improved pretreatment in clogging mat formation may need to be qualified in view of other research studies. It has also shown that effluent quality may be a factor in the case of highly permeable soils but this may not be the case with fine structured soils. The ultimate test of onsite sewage treatment system efficiency rests with the final disposal of effluent. The implication of system failure as evidenced from the surface ponding of effluent or the seepage of contaminants into the groundwater can be very serious as it can lead to environmental and public health impacts. Significant microbial contamination of surface and groundwater has been attributed to septic tank effluent. There are a number of documented instances of septic tank related waterborne disease outbreaks affecting large numbers of people. In a recent incident, the local authority was found liable for an outbreak of viral hepatitis A and not the individual septic tank owners as no action had been taken to remedy septic tank failure. This illustrates the responsibility placed on local authorities in terms of ensuring the proper operation of onsite sewage treatment systems. Even a properly functioning soil absorption system is only capable of removing phosphorus and microorganisms. The nitrogen remaining after plant uptake will not be retained in the soil column, but will instead gradually seep into the groundwater as nitrate. Conditions for nitrogen removal by denitrification are not generally present in a soil absorption bed. Dilution by groundwater is the only treatment available for reducing the nitrogen concentration to specified levels. Therefore based on subsurface conditions, this essentially entails a maximum allowable concentration of septic tanks in a given area. Unfortunately nitrogen is not the only wastewater pollutant of concern. Relatively long survival times and travel distances have been noted for microorganisms originating from soil absorption systems. This is likely to happen if saturated conditions persist under the soil absorption bed or due to surface runoff of effluent as a result of system failure. Soils have a finite capacity for the removal of phosphorus. Once this capacity is exceeded, phosphorus too will seep into the groundwater. The relatively high mobility of phosphorus in sandy soils have been noted in the literature. These issues have serious implications in the design and siting of soil absorption systems. It is not only important to ensure that the system design is based on subsurface conditions but also the density of these systems in given areas is a critical issue. This essentially involves the adoption of a land capability approach to determine the limitations of an individual site for onsite sewage disposal. The most limiting factor at a particular site would determine the overall capability classification for that site which would also dictate the type of effluent disposal method to be adopted.

Influence of filtration on I/O particle concentration ratios at urban office buildings

Epidemiological research has consistently shown an association between fine and ultrafine particle concentrations, and increases in both respiratory and cardiovascular morbidity and mortality. These particles, often found in vehicle emissions outside buildings, can penetrate inside via their envelopes and mechanically ventilated systems. Indoor activities such as printing, cooking and cleaning, as well as the movement of building occupants are also an additional source of these particles. In this context, the filtration systems of mechanically ventilated buildings can reduce indoor particle concentrations. Several studies have quantified the efficiency of dry-media and electrostatic filters, but they mainly focused on the particle size range > 300 nm. Some others studied ultrafine particles but their investigations were conducted in laboratories. At this point, there is still only limited information on in situ filter efficiency and an incomplete understanding of filtration influence on I/O ratios of particle concentrations. To help address these gaps in knowledge and provide new information for the selection of appropriate filter types in office building HVAC systems, we aimed to: (1) measure particle concentrations at up and down stream flows of filter devices, as well as outdoor and indoor office buildings; (2) quantify efficiency of different filter types at different buildings; and (3) assess the impact of these filters on I/O ratios at different indoor and outdoor source operation scenarios.

Performance assessment of exponential Rosenbrock method for large systems of ODE

This paper studies time integration methods for large stiff systems of ordinary differential equations (ODEs) of the form u'(t) = g(u(t)). For such problems, implicit methods generally outperform explicit methods, since the time step is usually less restricted by stability constraints. Recently, however, explicit so-called exponential integrators have become popular for stiff problems due to their favourable stability properties. These methods use matrix-vector products involving exponential-like functions of the Jacobian matrix, which can be approximated using Krylov subspace methods that require only matrix-vector products with the Jacobian. In this paper, we implement exponential integrators of second, third and fourth order and demonstrate that they are competitive with well-established approaches based on the backward differentiation formulas and a preconditioned Newton-Krylov solution strategy.

Detecting approximate clones in process model repositories with Apromore

Approximate clone detection is the process of identifying similar process fragments in business process model collections. The tool presented in this paper can efficiently cluster approximate clones in large process model repositories. Once a repository is clustered, users can filter and browse the clusters using different filtering parameters. Our tool can also visualize clusters in the 2D space, allowing a better understanding of clusters and their member fragments. This demonstration will be useful for researchers and practitioners working on large process model repositories, where process standardization is a critical task for increasing the consistency and reducing the complexity of the repository.

An investigation into factors affecting electron density calibration for a megavoltage cone-beam CT system

There is a growing interest in the use of megavoltage cone-beam computed tomography (MV CBCT) data for radiotherapy treatment planning. To calculate accurate dose distributions, knowledge of the electron density (ED) of the tissues being irradiated is required. In the case of MV CBCT, it is necessary to determine a calibration-relating CT number to ED, utilizing the photon beam produced for MV CBCT. A number of different parameters can affect this calibration. This study was undertaken on the Siemens MV CBCT system, MVision, to evaluate the effect of the following parameters on the reconstructed CT pixel value to ED calibration: the number of monitor units (MUs) used (5, 8, 15 and 60 MUs), the image reconstruction filter (head and neck, and pelvis), reconstruction matrix size (256 by 256 and 512 by 512), and the addition of extra solid water surrounding the ED phantom. A Gammex electron density CT phantom containing EDs from 0.292 to 1.707 was imaged under each of these conditions. The linear relationship between MV CBCT pixel value and ED was demonstrated for all MU settings and over the range of EDs. Changes in MU number did not dramatically alter the MV CBCT ED calibration. The use of different reconstruction filters was found to affect the MV CBCT ED calibration, as was the addition of solid water surrounding the phantom. Dose distributions from treatment plans calculated with simulated image data from a 15 MU head and neck reconstruction filter MV CBCT image and a MV CBCT ED calibration curve from the image data parameters and a 15 MU pelvis reconstruction filter showed small and clinically insignificant differences. Thus, the use of a single MV CBCT ED calibration curve is unlikely to result in any clinical differences. However, to ensure minimal uncertainties in dose reporting, MV CBCT ED calibration measurements could be carried out using parameter-specific calibration measurements.

A general model for MAC generation using direct injection

This paper presents a model for generating a MAC tag by injecting the input message directly into the internal state of a nonlinear filter generator. This model generalises a similar model for unkeyed hash functions proposed by Nakano et al. We develop a matrix representation for the accumulation phase of our model and use it to analyse the security of the model against man-in-the-middle forgery attacks based on collisions in the final register contents. The results of this analysis show that some conclusions of Nakano et al regarding the security of their model are incorrect. We also use our results to comment on several recent MAC proposals which can be considered as instances of our model and specify choices of options within the model which should prevent the type of forgery discussed here. In particular, suitable initialisation of the register and active use of a secure nonlinear filter will prevent an attacker from finding a collision in the final register contents which could result in a forged MAC.

Prevalence of papillary changes and folliculosis of the palpebral conjunctiva in asymptomatic Chinese children

Purpose To determine the prevalence of papillary changes of the upper palpebral conjunctiva and folliculosis of the lower palpebral conjunctiva in Chinese children with no history of contact lens wear. Method Ninety-nine subjects (aged 6–15 years old) who were interested in a myopia control study were screened for papillary changes and folliculosis of the palpebral conjunctiva. Photodocumentation was performed under white and blue light (after the application of fluorescein) with a yellow filter and the photographs were graded by a group of practitioners according to a pre-set grading scale. Analysis was performed with the subjects divided into groups according to gender and age. Results More than 48% of the subjects had clinically significant (≥Grade 3) papillary changes in the upper palpebral conjunctiva. The prevalence of significant folliculosis in the lower lid was about 33%. The prevalence of significant papillary changes and folliculosis were similar between genders. No differences were observed between younger (age ≤ 10 years old) and older (age > 10 years old) in papillary changes but younger subjects showed a higher prevalence of folliculosis. Conclusions The prevalences of clinically significant papillary changes and folliculosis of unknown aetiology are high in Chinese children.

Adaptive stepsize based on control theory for stochastic differential equations

The numerical solution of stochastic differential equations (SDEs) has been focused recently on the development of numerical methods with good stability and order properties. These numerical implementations have been made with fixed stepsize, but there are many situations when a fixed stepsize is not appropriate. In the numerical solution of ordinary differential equations, much work has been carried out on developing robust implementation techniques using variable stepsize. It has been necessary, in the deterministic case, to consider the "best" choice for an initial stepsize, as well as developing effective strategies for stepsize control-the same, of course, must be carried out in the stochastic case. In this paper, proportional integral (PI) control is applied to a variable stepsize implementation of an embedded pair of stochastic Runge-Kutta methods used to obtain numerical solutions of nonstiff SDEs. For stiff SDEs, the embedded pair of the balanced Milstein and balanced implicit method is implemented in variable stepsize mode using a predictive controller for the stepsize change. The extension of these stepsize controllers from a digital filter theory point of view via PI with derivative (PID) control will also be implemented. The implementations show the improvement in efficiency that can be attained when using these control theory approaches compared with the regular stepsize change strategy.

Image Region Salience for Improving Appearance-Based Place Recognition using a Supervised Classifier System

Many state of the art vision-based Simultaneous Localisation And Mapping (SLAM) and place recognition systems compute the salience of visual features in their environment. As computing salience can be problematic in radically changing environments new low resolution feature-less systems have been introduced, such as SeqSLAM, all of which consider the whole image. In this paper, we implement a supervised classifier system (UCS) to learn the salience of image regions for place recognition by feature-less systems. SeqSLAM only slightly benefits from the results of training, on the challenging real world Eynsham dataset, as it already appears to filter less useful regions of a panoramic image. However, when recognition is limited to specific image regions performance improves by more than an order of magnitude by utilising the learnt image region saliency. We then investigate whether the region salience generated from the Eynsham dataset generalizes to another car-based dataset using a perspective camera. The results suggest the general applicability of an image region salience mask for optimizing route-based navigation applications.

Fourier Lucas-Kanade Algorithm

In this paper we propose a framework for both gradient descent image and object alignment in the Fourier domain. Our method centers upon the classical Lucas & Kanade (LK) algorithm where we represent the source and template/model in the complex 2D Fourier domain rather than in the spatial 2D domain. We refer to our approach as the Fourier LK (FLK) algorithm. The FLK formulation is advantageous when one pre-processes the source image and template/model with a bank of filters (e.g. oriented edges, Gabor, etc.) as: (i) it can handle substantial illumination variations, (ii) the inefficient pre-processing filter bank step can be subsumed within the FLK algorithm as a sparse diagonal weighting matrix, (iii) unlike traditional LK the computational cost is invariant to the number of filters and as a result far more efficient, and (iv) this approach can be extended to the inverse compositional form of the LK algorithm where nearly all steps (including Fourier transform and filter bank pre-processing) can be pre-computed leading to an extremely efficient and robust approach to gradient descent image matching. Further, these computational savings translate to non-rigid object alignment tasks that are considered extensions of the LK algorithm such as those found in Active Appearance Models (AAMs).

On robust face recognition via sparse encoding : the good, the bad, and the ugly

In the field of face recognition, Sparse Representation (SR) has received considerable attention during the past few years. Most of the relevant literature focuses on holistic descriptors in closed-set identification applications. The underlying assumption in SR-based methods is that each class in the gallery has sufficient samples and the query lies on the subspace spanned by the gallery of the same class. Unfortunately, such assumption is easily violated in the more challenging face verification scenario, where an algorithm is required to determine if two faces (where one or both have not been seen before) belong to the same person. In this paper, we first discuss why previous attempts with SR might not be applicable to verification problems. We then propose an alternative approach to face verification via SR. Specifically, we propose to use explicit SR encoding on local image patches rather than the entire face. The obtained sparse signals are pooled via averaging to form multiple region descriptors, which are then concatenated to form an overall face descriptor. Due to the deliberate loss spatial relations within each region (caused by averaging), the resulting descriptor is robust to misalignment & various image deformations. Within the proposed framework, we evaluate several SR encoding techniques: l1-minimisation, Sparse Autoencoder Neural Network (SANN), and an implicit probabilistic technique based on Gaussian Mixture Models. Thorough experiments on AR, FERET, exYaleB, BANCA and ChokePoint datasets show that the proposed local SR approach obtains considerably better and more robust performance than several previous state-of-the-art holistic SR methods, in both verification and closed-set identification problems. The experiments also show that l1-minimisation based encoding has a considerably higher computational than the other techniques, but leads to higher recognition rates.