Table 4 Attribute names and level classification.

From: Comprehensive walkability assessment of urban pedestrian environments using big data and deep learning techniques

Attribute name

Description

Calculation formula

Explanation

Classification standard

Greenery visibility rate

The influence of trees, shrubs, and other greenery on pedestrians’ psychological feelings.

\(\:{P}_{green,\:\:j}=\frac{{{\sum\:}_{i=1}^{4}GP}_{i}}{{\sum\:}_{i=1}^{4}{P}_{i}}\times\:100\%\)

\(\:{P}_{green,\:\:j}\) represents the greenery visibility rate at collection point j. \(\:G{P}_{i}\:\)is the green pixel value in the i-th image, and \(\:{P}_{i}\)​ is the total pixel value of the image.

Classified according to Zhu et al.84: Low (\(P_{{green,j}} \; < \;5\%\)), Medium (\(5\% \; \le \;P_{{green,j}} \; < \;35\%\)), High (\(P_{{green,j}} \; \ge \;35\%\)).

Interface transparency

The sense of enclosure created by buildings and walls, affecting pedestrians’ visual transparency and spatial perception.

\(\:{P}_{tran,\:\:j}=\frac{{{\sum\:}_{i=1}^{4}BP}_{i}+W{P}_{i}}{{\sum\:}_{i=1}^{4}{P}_{i}}\times\:100\%\)

\(\:{P}_{tran,\:\:j}\) represents the interface transparency at collection point\(\:j\). \(\:B{P}_{i}\) and \(\:W{P}_{i}\)are the building and wall pixel values in the i-th image, respectively, and, \(\:{P}_{i}\)is the total pixel value of the image.

Referring to Zhang et al.85 with slight adjustments: Open(\(P_{{green,j}} \; < \;5\%\)), Semi-open (\(5\% \; \le \;P_{{green,j}} \; < \;40\%\)), Closed (\(P_{{green,j}} \; \ge \;40\%\)).

Relative sidewalk width

The ratio of sidewalk width to roadway width, affecting pedestrians’ walking comfort and sense of safety.

\(\:{P}_{width,j}=\frac{{\sum\:}_{i=1}^{4}{SP}_{i}}{{\sum\:}_{i=1}^{4}{RP}_{i}}\times\:100\%\)

\(\:{P}_{width,\:\:j}\) represents the relative sidewalk width at collection point \(\:j.\)\(\:S{P}_{i}\:\)is the sidewalk pixel value in the i-th image, and, \(\:{RP}_{i}\) is the roadway pixel value.

Narrow(\(P_{{width,j}} \; < \;1.5\%\)), Medium (\(1.5\% \; \le \;P_{{width,j}} \; < 25\%\)), Wide (\(P_{{width,j}} \; \ge 25\%\)).

Barrier separation

Separation facilities between sidewalks and roadways, directly impacting pedestrian safety and psychological security.

\(\:{P}_{fence,\:\:j}=\frac{{{\sum\:}_{i=1}^{4}FP}_{i}}{{\sum\:}_{i=1}^{4}{P}_{i}}\times\:100\%\)

\(\:{P}_{fence,\:\:j}\) represents the barrier pixel ratio at collection point j. \(\:F{P}_{i}\) is the barrier pixel value in the i-th image, and \(\:{P}_{i}\) is the total pixel value of the image.

None(\(P_{{fence,j}} \; < \;5\%\)), Present (\(P_{{fence,j}} \; \ge \;5\%\)).

Motor vehicle flow

The number of motor vehicles such as cars and buses, determining pedestrian noise exposure and air quality experience.

\(\:{N}_{traffic,\:\:j}=\sum\:_{i=1}^{4}T{P}_{i}\)

\(\:{N}_{traffic,\:\:j}\) represents the total number of motor vehicles at collection point j. \(\:T{P}_{i}\) is the number of motor vehicles in the i-th image.

Classified by quartiles: Low (N less than the first quartile), Medium (N between the first and second quartiles), High (N greater than the second quartile)

Pedestrian flow

The number of pedestrians on the street, affecting interaction between pedestrians and personal space.

\(\:{N}_{pedestrian,\:\:j}=\sum\:_{i=1}^{4}P{P}_{i}\)

\(\:{N}_{pedestrian,\:\:j}\) represents the total number of pedestrians at collection point j. \(\:P{P}_{i}\) is the number of pedestrians in the i-th image.

Street facilities

The number of street elements such as signs, billboards, and benches, significantly influencing street cognition.

\(\:{N}_{facilities,\:\:j}=\sum\:_{i=1}^{4}L{P}_{i}\)

\(\:{N}_{facilities,\:\:j}\) represents the total number of street facilities at collection point j. \(\:L{P}_{i}\) is the number of street facilities in the i-th image.