Traffic and Road Sign Recognition

This thesis presents a system to recognise and classify road and traffic signs for the purpose of developing an inventory of them which could assist the highway engineers’ tasks of updating and maintaining them. It uses images taken by a camera from a moving vehicle. The system is based on three major stages: colour segmentation, recognition, and classification. Four colour segmentation algorithms are developed and tested. They are a shadow and highlight invariant, a dynamic threshold, a modification of de la Escalera’s algorithm and a Fuzzy colour segmentation algorithm. All algorithms are tested using hundreds of images and the shadow-highlight invariant algorithm is eventually chosen as the best performer. This is because it is immune to shadows and highlights. It is also robust as it was tested in different lighting conditions, weather conditions, and times of the day. Approximately 97% successful segmentation rate was achieved using this algorithm.Recognition of traffic signs is carried out using a fuzzy shape recogniser. Based on four shape measures - the rectangularity, triangularity, ellipticity, and octagonality, fuzzy rules were developed to determine the shape of the sign. Among these shape measures octangonality has been introduced in this research. The final decision of the recogniser is based on the combination of both the colour and shape of the sign. The recogniser was tested in a variety of testing conditions giving an overall performance of approximately 88%.Classification was undertaken using a Support Vector Machine (SVM) classifier. The classification is carried out in two stages: rim’s shape classification followed by the classification of interior of the sign. The classifier was trained and tested using binary images in addition to five different types of moments which are Geometric moments, Zernike moments, Legendre moments, Orthogonal Fourier-Mellin Moments, and Binary Haar features. The performance of the SVM was tested using different features, kernels, SVM types, SVM parameters, and moment’s orders. The average classification rate achieved is about 97%. Binary images show the best testing results followed by Legendre moments. Linear kernel gives the best testing results followed by RBF. C-SVM shows very good performance, but ?-SVM gives better results in some case.

Utvärdering av digitala provtryckssystem

Six different Digital Proofing Systems from three different techniques have been evaluated as totechnique, printing quality, economy and usability. Digital proof from two paper qualities, coatedand uncoated, has been compared with references printed in offset, to see how good they match eachother. Only two Proofing Systems manage to print on reference paper. The other Proofing Systemsuse special paper for digital proof.Measurements and visuell judgement show that the Digital Proofing Systems visualise referencepictures with quite good quality. Proof optimised for coated paper visualise the colours with goodresult. Proof optimised for uncoated paper shows higher quality than the references, which depends onthe surface of the proofing paper. Comparison between reference paper and proofing paper has takenplace as to differences in colour and paper quality.The Digital Proofing Systems are fully automatic, which demand a quite comprised education forcorrect handling. The purchase price and printing costs vary considerably between the ProofingSystems.

Layout och typografi i vetenskapliga artiklar : En kartläggning av konventioner gällande grafisk formgivning inom matematik, medicin och utbildningsvetenskap

En deskriptiv visuell innehållsanalys genomfördes med en korpus bestående av vetenskapliga artiklar från 30 topprankade tidskrifter jämnt fördelade på de tre disciplinerna matematik, medicin och utbildningsvetenskap, med syfte att undersöka om det fanns ett samband mellan tidskrifternas disciplintillhörighet och den grafiska formgivningen. Kartläggningen omfattade layout (bl.a. sidformat, spalter, marginalstorlek, användning av grafiska element, placering av tabeller, figurer och pagina) och typografiskt utseende hos titel, rubriker, brödtext och abstrakt (t.ex. teckenstorlek, textjustering, teckengrovlek, typsnitt, färg och radavstånd). Resultaten visar att disciplinerna använde olika men även gemensamma konventioner i sitt grafiska utseende. Matematik uppvisade enkelspaltig layout med liten typografisk variation, genomgående användning av typsnitt från familjen Transitional eller Transitional/Didone för hela dokumentet, avsaknad av kulört färg och sällan användning av grafiska element. Medicin uppvisade större variation med användning av kulört färg, icke-centrerad typografi med både sansseriff- och seriff-typsnitt, grafiska element och ramar som förstärker det ofta använda dubbelspaltiga gridsystemet. Utbildningsvetenskap uppvisade en grafisk design som återfanns i båda disciplinerna, men påminde mest om matematik till utseendet. Kodbok ingår i sin helhet som bilaga. Uppsatsen i sig är delvis formgiven med utgångspunkt i de erhållna resultaten.

Printing colour hard proofs using EFI Colorproof XF v. 3.1 and Photoshop CS3, and production substrates.

EFI Colorproof XF was found to be more convenient from a user’s aspect, and had features which are covered in the ISO 12647-7 standard (e.g. the ability to simulate screening and print margin information), which Photoshop CS3 lacked. None of the proofing systems distinguished itself in a clear way from the other; sometimes, on certain substrates, Photoshop CS3 produced most accurate colours, sometimes EFI Colorproof XF did. Further investigations need to be carried out to tell more exactly which system produce most accurate colours. Only 6 out of 34 simulation-combinations had colours within the tolerances in the standard. The result also shows that the production substrates should not be used as proofing substrates. Instead the proofing papers especially made for ink jet should be used to obtain more colour-accurate prints.