THE ROLE OF HILL-SHADING
IN TOURIST MAPS:
an eye-tracking study
www.eyetracking.upol.cz
Stanislav POPELKA
Department ...
EVALUATION OF 3D VISUALIZATION IN
GIS WITH THE USE OF EYE-TRACKING
• 3D visualization in three areas:
• 3D visualizatio...
Our laboratory
• Equipment:
– SMI RED 250
– Sampling frequency 120 Hz
– Attached web camera
• Software
– SMI Experim...
Shading in maps
• Three categories of shading in cartography
– Slope Shading
– Oblique shading
– Combined shading
• M...
The influence of lighting from the South-East
• Task – mark the highest point in the map
www.eyetracking.upol.cz
Perspe...
The role of hill-shading in tourist maps
• Eye-tracking experiment
• Total of 40 respondents (Within – subject study)
•...
Stimuli in the
experiment
www.eyetracking.upol.cz
Results of the questionnaire
• After the end of experiment, respondents
were asked to fill the questionnaire:
– Which t...
Visualization of eye-tracking data
Attention maps in the stimuli from task 3 created in OGAMA software. Kernel 100px
Ide...
Visualization of eye-tracking data
• Visualization of the grid representing the
distribution of fixations within the sti...
www.eyetracking.upol.cz
Analyses of eye-tracking metrics
• At the beginning, the Trial Duration was analyzed
„ How long it took to answer“
Comp...
Analyses of eye-tracking metrics
• Fixation Count - A larger number of fixations indicates a lower level of efficiency
i...
Analyses of eye-tracking metrics
• Data were tested using the Wilcoxon test
• Nonparametric test – data had not a normal...
Analyses of eye-tracking metrics
• In the next step, data were analyzed for all maps together
(shaded vs. non-shaded)
•...
Analyses of the role of the task
• The last part of data analyses was evaluation of the role of the
task
„If respondent...
www.eyetracking.upol.cz
Analyses of the role of the task
• Finally, the statistical dependence between the type of the task (village –
hill) and...
Dedendance between task and visualization method
• Type of the task:
„It was more difficult to find a village, especiall...
Conclusion
• An eye-tracking experiment containing tourist maps with and without
shading was created
• The questionnair...
References
• BERNABÉ-POVEDA, M. A., SÁNCHEZ-ORTEGA, I., ÇÖLTEKIN, A. Techniques for Highlighting Relief on
Orthoimaginer...
Thank you for your attention
The presentation was made with the support of European Social Fund - Operational Program
Ed...
of 22

Popelka - The role of hill-shading in tourist maps: an eye-tracking study

Presentation from Third InDOG Doctoral Conference in Olomouc, Czech Republic. 13. - 16. October 2014
Published on: Mar 4, 2016
Published in: Data & Analytics      
Source: www.slideshare.net


Transcripts - Popelka - The role of hill-shading in tourist maps: an eye-tracking study

  • 1. THE ROLE OF HILL-SHADING IN TOURIST MAPS: an eye-tracking study www.eyetracking.upol.cz Stanislav POPELKA Department of geoinformatics Palacký University in Olomouc Czech Republic
  • 2. EVALUATION OF 3D VISUALIZATION IN GIS WITH THE USE OF EYE-TRACKING • 3D visualization in three areas: • 3D visualization of terrain – Perspective maps (Popelka and Brychtová, 2013) – Hill-shading in maps • 3D visualization of cities – Photorealistic visualization (Popelka and Dědková, 2013) – Non-photorealistic visualization (Popelka and Doležalová, 2014) • Thematic 3D maps www.eyetracking.upol.cz
  • 3. Our laboratory • Equipment: – SMI RED 250 – Sampling frequency 120 Hz – Attached web camera • Software – SMI Experiment center – SMI BeGaze – OGAMA – V-Analytics (a.k.a. CommonGIS) – R www.eyetracking.upol.cz
  • 4. Shading in maps • Three categories of shading in cartography – Slope Shading – Oblique shading – Combined shading • Most frequently used is Conventional lighting Bernabé-Poveda et al., 2011 – The light is coming from the upper left corner of the map – This situation cannot occur in nature, but is the most effective for the perception of plasticity (Imhof, 2007) – Several reasons for support of South-East shading exists – BUT…users are so completely conditioned to a light source from above left that they will percieve the terrain as a negativ (Imhof, 2007) www.eyetracking.upol.cz
  • 5. The influence of lighting from the South-East • Task – mark the highest point in the map www.eyetracking.upol.cz Perspective 3D map 2D map illuminated from the bottom-right corner
  • 6. The role of hill-shading in tourist maps • Eye-tracking experiment • Total of 40 respondents (Within – subject study) • Total of 12 stimuli (screenshots from Mapy.cz) – always a pair of maps • The task was to find and mark particular hill or village www.eyetracking.upol.cz
  • 7. Stimuli in the experiment www.eyetracking.upol.cz
  • 8. Results of the questionnaire • After the end of experiment, respondents were asked to fill the questionnaire: – Which type of map was more usable for solving the task – Which map type they like more www.eyetracking.upol.cz
  • 9. Visualization of eye-tracking data Attention maps in the stimuli from task 3 created in OGAMA software. Kernel 100px Identified fixations in the stimuli from task 6 – Created from data of all respondents in OGAMA software. Radius = 20 www.eyetracking.upol.cz
  • 10. Visualization of eye-tracking data • Visualization of the grid representing the distribution of fixations within the stimulus Ooms et al., 2014 • A regular grid of size 5x10 cells was created above each stimulus • These values can be used in further analyses (adding, substracting, etc…) www.eyetracking.upol.cz
  • 11. www.eyetracking.upol.cz
  • 12. Analyses of eye-tracking metrics • At the beginning, the Trial Duration was analyzed „ How long it took to answer“ Comparing of the median values of Trial Duration. Error bars represent the interquartile range. www.eyetracking.upol.cz
  • 13. Analyses of eye-tracking metrics • Fixation Count - A larger number of fixations indicates a lower level of efficiency in the search. Goldberg and Kotval, 1999 • Scanpath Length - Depending on the size of this metric comprehensibility of stimuli or difficulty of the questions can be derived. Goldberg and Helfman, 2011 www.eyetracking.upol.cz
  • 14. Analyses of eye-tracking metrics • Data were tested using the Wilcoxon test • Nonparametric test – data had not a normal distribution • Statisticaly significant differences between shaded and non-shaded stimuli are highlighted Trial Duration Fixation Count Fixation Duration Scanpath Length Trial 1 0.06856 0.05789 0.65 0.01411 Trial 2 0.03783 0.03093 0.004077 0.01853 Trial 3 0.02155 0.02791 0.08049 0.01925 Trial 4 0.02989 0.03659 0.3567 0.02497 Trial 5 0.8785 0.9222 0.3303 0.8368 Trial 6 0.000009 0.0004388 0.07838 0.000008 Results of the testing of the difference between shaded and non-shaded maps for four evaluated eye-tracking metrics. Data were analyzed using the Wilcoxon test for each pair of stimuli separately. Statistically significant differences (significance level of α = 0.05) are highlighted . www.eyetracking.upol.cz
  • 15. Analyses of eye-tracking metrics • In the next step, data were analyzed for all maps together (shaded vs. non-shaded) • Statistically significant differences were found for all evaluated eye-tracking metrics except Fixation Duration www.eyetracking.upol.cz
  • 16. Analyses of the role of the task • The last part of data analyses was evaluation of the role of the task „If respondents were trying to find village (trial 1-3) or the hill (trial 4-6)“ • Statisticaly significant differences were found for all evaluated eye-tracking metrics Village vs. Hill Aplha V p-value statement Trial Duration 0.05 5708.5 4.369e-16 Rejecting H0 Fixation Count 0.05 4954 <2.2e-16 Rejecting H0 Fixation Duration 0.05 17794.5 1.979e-06 Rejecting H0 Scanpath Length 0.05 5499 <2.2e-16 Rejecting H0 Results of the testing of the difference between task dealing with finding village (trial 1-3) and hill (task 4-6) for four evaluated eye-tracking metrics. Data were analyzed using the Wilcoxon test. Statistically significant differences (significance level of α = 0.05) are highlighted . www.eyetracking.upol.cz
  • 17. www.eyetracking.upol.cz
  • 18. Analyses of the role of the task • Finally, the statistical dependence between the type of the task (village – hill) and type of visualization (shaded – non-shaded) was analysed • Testing of Trial Duration metric • Two-way ANOVA + TukeyHSD Dvojice Diff Lwr Upr P adj 3D hill-2D hill 655.050 -1828.5881 3138.688 0.9046902 2D village - 2D hill 4088.767 1605.1286 6572.405 0.0001544 3D village - 2D hill 8169.833 5686.1952 10653.471 0.0000000 2D village - 3D hill 3433.717 950.0786 5917.355 0.0022643 3D village - 3D hill 7514.783 5031.1452 9998.421 0.0000000 3D village - 2D village 4081.067 1597.4286 6564.705 0.0001598 Results of the comparison of the effect of question (village - hill) and visualization (shaded – non-shaded) for eye-tracking metric Trial Duration. (ANOVA* + TukeyHSD) *Residuas from ANOVA comparison had not a normal distribution, but result of Bootstraping confirmed the accuracy of the results www.eyetracking.upol.cz
  • 19. Dedendance between task and visualization method • Type of the task: „It was more difficult to find a village, especially in the map with the shading • Type of the visualization method: „Shaded variant of the maps were less usable – especially when the task was to find a village“ www.eyetracking.upol.cz
  • 20. Conclusion • An eye-tracking experiment containing tourist maps with and without shading was created • The questionnaire showed that users liked shaded maps more than non-shaded. • Shaded vs. Non-shaded – When analysing eye-tracking metrics Trial Duration, Fixation Count a Scanpath Length statisticaly significant differences were found. – Higher values were observed in the case of shaded maps • Village vs. Hill – Statisticaly significant differences were found for all analysed eye-tracking metrics – Higher values were observed in the case of task about finding a village (except Fixation Duration) • It was easier to find hill than village • Search in shaded maps was less effective. www.eyetracking.upol.cz
  • 21. References • BERNABÉ-POVEDA, M. A., SÁNCHEZ-ORTEGA, I., ÇÖLTEKIN, A. Techniques for Highlighting Relief on Orthoimaginery. Procedia-Social and Behavioral Sciences, 2011, 21, pp. 346-352. • GOLDBERG, J., HELFMAN, J. Eye tracking for visualization evaluation: Reading values on linear versus radial graphs. Information Visualization, 2011, 10(3), pp. 182-195. • GOLDBERG, J. H., KOTVAL, X. P. Computer interface evaluation using eye movements: methods and constructs. International Journal of Industrial Ergonomics, 1999, 24(6), pp. 631-645. • IMHOF, E. Cartographic relief presentation. ESRI, Inc., 2007. • OOMS, K., DE MAEYER, P., FACK, V. Study of the attentive behavior of novice and expert map users using eye tracking. Cartography and Geographic Information Science, 2014, 41(1), pp. 37-54. • POPELKA, S., BRYCHTOVA, A. Eye-tracking Study on Different Perception of 2D and 3D Terrain Visualisation. Cartographic Journal, Aug 2013, 50(3), pp. 240-246. • POPELKA, S., DEDKOVA, P. Extinct Village 3D Visualization and its Evaluation with Eye-Movement Recording. ICCSA 2014 - Lecture Notes in Computer Science Volume 8579 Springer International Publishing, 2014, 10p. • POPELKA, S., DOLEŽALOVÁ, J. Non-photorealistic 3D Visualization in City Maps: An Eye-tracking Study, New Trends in Cartography, 2014, 12p. - in print www.eyetracking.upol.cz
  • 22. Thank you for your attention The presentation was made with the support of European Social Fund - Operational Program Education for Competitiveness (project CZ.1.07/2.2.00/28.0078 „InDOG“) and Internal Grant Agency of Palacký University (project: PrF_2013_024) www.eyetracking.upol.cz standa.popelka@gmail.com

Related Documents