Plant color was an important landscape of the urban green space, and quantitative analysis of its characteristics could provide a basis for plant color configuration. The leaf color attributes of 80 colored-leaf plants were studied and the Scenic Beauty Estimates (SBE) was used to evaluate their beauty. The data of leaves color were obtained by MATLAB software and scanner under the hue–saturation–brightness (HSB) system and then were further analyzed to determine the relationship between the SBE and color attributes (HSB). Results showed that the leaf colors of 80 plants contained 13 hues in total, which belonged to six color systems. The species contained in different color systems, from most to least, were yellow–green series (27 species) > orange series (19 species) > red series (18 species) > yellow series (10 species) > blue-green (four species) > purple (two species). There were fewer blue and purple series species. The brightness and saturation of color were mainly distributed in the middle (26–50%), middle and high (51–75%) stages, and those of the yellow–green system belonged to the medium and high stages and were significantly higher than other color systems. There is a significant positive correlation between the brightness and saturation of color and the value of SBE, and an SBE evaluation model was established based on the color attributes. The results are helpful to select plants with higher beauty and create a rich plant color landscape.
In urban green space, rich colors were an important part of the landscape, which could first attract people’s attention. The colors of garden plants mainly include leaf color, flower color, and fruit color . Flowers and fruits only appeared during a certain period, but the leaf colors were the main feature to define the type of garden plants color and landscape color and had more research value .
The research on color plants was widely covered in urban planning, garden architecture, and landscaping. Badami investigated the innovative experimentation in Italy about the compilation of color plants and presented the case of the color plan for the Egadi Islands . Yuning illustrated the value and significance of the quantitative expression in the landscape color research on the color of the Ming Dynasty City Wall in Nanjing . Other research on plant color included factors affecting color presentation , quantification of the color [6,7], and the changes in leaf color in the four seasons. Colored-leaf plants include two types: one in which the leaf color changes in different seasons, and another in which leaf color can maintain one color other than green throughout the growing season [8,9]. In this article, we chose the latter for research, because of their high ornamental value of leaf color throughout the growth period [10,11,12]. At present, most research on colored-leaf plants focused on plant species, the mechanism of seasonal discoloration of leaves, and landscape application, and there was no quantitative analysis on the leaf color of colored-leaf plants [13,14,15]. The color sampling could be carried out by using color charts, colorimeters, and photography [16,17]. Taking color samples with color charts was too subjective; the workload of color measurement with a colorimeter was large; and the color obtained by the camera was greatly affected by the weather [18,19]. To avoid these shortcomings, in this study, the K-mean clustering was used to extract the main color from the photos, and leaf colors scanned were used as an aid in special cases to determine the data of color, and then the data were transformed into hue–saturation–brightness (HSB) data. The HSB color mode is a three-channel color analysis technology based on HSB, which is close to the human eye’s sense of color . Then, a psychophysical evaluation Scenic Beauty Estimates (SBE) method proposed by Daniel and Boster in 1976 was used to evaluate the preference of the public for plant color [21,22].
Finally, a quantitative analysis of the color of leaves was conducted; the plants with a high SBE score were selected, and all of them should be used in the color configuration of gardens to create a colorful landscape.
2 Materials and methods
2.1 Materials selection and picture shooting
We first determined the preliminary list of tree species based on the practical application of color-leafed plants in the books “the World Encyclopedia of Garden Plants and Flowers” and “the Production, Cultivation and Application of Color Leafed Plants Compiled” and then selected the specific sample trees by investigating the use of them across the country. We chose tree species with a frequency of more than five in the city, which meant that they were commonly used. Finally, we selected 80 common colored-leaf plants (Appendix I) in public green space as the research objects, including trees, shrubs, vines, and groundcover plants. At least five healthy individuals of each plant were chosen as samples for photography and leaf scanning as five replicates. Eighty colored-leaf plants belonged to 35 families and 56 genera, including 36 species of trees, 21 species of shrubs, 22 species of herbs, and 1 species of vine (Table 1).
|Types||Number of families (genera and species)||Family (number) name||Life form||Number of species|
|Families with 10 or more species||2 (8, 23)||Rosaceae (13) and Aceraceae (10)||Arbor||Evergreen arbor (35)|
|Deciduous arbor (1)|
|Families with 5–9 species||3 (12, 17)||Fabaceae (6), Poaceae (5), and Cupressaceae (5)||Shrub||Evergreen shrub (12)|
|Deciduous shrub (9)|
|Families with 2–4 species||4 (10, 15)||Oleaceae (4), Labiatae (4), Ulmaceae (3), Liliaceae (2), and lacqueraceae (2)||Vine||1|
|Families with only one species||25 (25, 25)||Oxaliaceae, Ginkgoaceae, Commelinaceae, Convolvulaceae, Berberidaceae, Amaranthaceae, Myrtaceae, Pinaceae, Caryophyllaceae, Taxodiaceae, Cyperaceae, Moraceae, Loniceraceae, Begoniaceae, Lythrymaceae, Canniaceae, Verbenaceae, Salicaceae, Fagaceae, Compositae, Crassulaceae, Hamamelidaceae, Bupleurumceae, Saxifragaceae, and Primulaceae is a single family, genus and species||Herb||Perennial herb (21)|
|Annual herb (1)|
The leaf color was stable from 10 June 2020 to 5 July 2020, with little change. We picked leaves from the upper, middle, and lower parts of the trees and then scanned them with an Epson V30 scanner. Scanned the white A4 paper and checked its HSB value as a correction before scanning the blades. The parameters of the scanner were as follows: the exposure type was set to documentation exposure, and the linear correction was selected as hue correction mode. Photos of the whole canopy of sample trees were taken by the camera (Sony A6000) at 10:00–14:00 on sunny days (shooting parameters were 14-bit RAW images at a spatial resolution of 6,000 × 4,000 Pixels, autofocus mode, aperture F8, iso100). About 2–3 photos were taken in each direction of sample trees for the next analysis. Throughout the shooting process, the same person used the same camera and kept the same height for shooting to minimize errors. Then, put a white A4 paper next to the object to be shot so that the difference caused by the environment could be corrected later.
2.2 Color extraction and data analysis
The main color of the leaves was extracted from the photos using K-means clustering in MATLAB software. It was necessary to use Photoshop CS6 software to make the photos' background white before clustering to eliminate the influence of external factors such as light. Compare the clustering results of the treated photos with the untreated photos and screen out the sample pictures with large differences.
K-means clustering steps:
Step 1: Selected a photo of the canopy and displayed each pixel in the color space. One photo has a finite number of N pixels, which are expressed as x i , i = 1, …, N, and its color information is x (R, G, B) − x i (R i , G i , B i ), and initialize K clustering centers C i (R i , G i , B i ).
where S m is all y points that meet the conditions in the three-dimensional region with radius h. k means that of n samples, and k points that meet the conditions are located in the S m region.
Step 2: With each C i (R i , G i , B i ) point as the center, calculated the sum of vector M(x) of each dimension in the region, and M(x) is the set of nearest point from x to C i .
Step 3: Determined the new center point with the mean value of M(x) to replace the original center.
Step 4: Repeated steps 2 and 3 until M(x) converged and took the color data of the final center point as the color data of the photo.
Step 5: Because the final color data is in red-green-blue (RGB) mode, it is necessary to convert the color value from RGB mode to HSB value according to the following formula:
For example, the K-means clustering results of Melaleuca bracteate and Loropetalum chinense var. rubrum are shown in Figure 1.
More than 70% of the colors of some plants could not be clustered because of their complex environmental colors, we used the mixed color method to obtain color data. Scanned leaves from different parts of sample trees to obtain their HSB, calculated the mean value of HSB in pairs, and then obtained the final HSB of leaf color through K-means clustering. We took Prunus × cistena ‘Pissardii as an example; the leaf color scanning results are shown in Figure 2; and the area of the red was the proportion of the main color.
According to the Ostwald color system, the hue (H) of a color was divided into 24 levels (Figure 3), and each hue accounts for 15° on the hue ring . Through K-means clustering, we got the HSB value of the color and used the H value to find the corresponding interval on the Ostwald hue circle and then drew the hue distribution ring (Figure 3). For example, the HSB of Prunus cerasifera ‘Newpiortii’ is 5°–61–36%, respectively, and the corresponding range of hue value 5° is 1–15°, which belongs to the red (H1). At the same time, the saturation (S) and brightness (B) were divided into low (0–25%), medium (26–50%), medium–high (51–75%), and high (76–100%) levels.
2.3 SBE of the color
The results of previous research showed that there was no significant difference between the landscape evaluation results of different types of evaluators . Therefore, teachers and students at the university were selected as evaluators. The canopy photos were randomly made into a slideshow, which was displayed to the evaluators using a projector in a classroom (correlated color temperature [CCT] is 5,000 K and luminance of lighting environment is 2,000 cd/m2). Because this method had a disadvantage that might affect the results, there was less chance to compare the colors of photos. Therefore, we briefly explained the evaluation criteria to the evaluators before starting the evaluation and played the slideshow quickly for the evaluators to give them an overall impression and a preliminary comparison of all photos, and then proceeded in turn, played slideshow. Each slide was held for 10 s, during which the evaluators needed to write down the score for each picture in turn on the evaluation form.
According to the evaluator’s preference for color, each photo was scored on a 10-point system (Table 2) in order to make the evaluation objective and true to reflect the evaluator’s feelings about the plant’s color and its ornamental effect. After that the evaluation data were gotten, and the SBE value was calculated based on these data. The calculation method was as follows: the frequency (f) of each value was counted to obtain its cumulative frequency (cf). We divided the cumulative frequency by the total number of evaluators (N) to obtain the cumulative probability (CP) and then used CP to find the single quantile of the normal distribution (Z). The SBE value of any photo was randomly selected and set to 0 as the benchmark value [25,26]. Then, we calculated the average of Z (MZ) of each group of photos, subtracted the MZ of the benchmark group from it, and multiplied their difference by 100 to obtain the final SBE value of that group. The calculation is as follows:
|9–10:||Superb||1. Please rate the pictures according to their preference for the color, ignoring other factors of the tree, such as tree shape, leaf shape, and all other factors except color|
|1–2||Disappointing||2. The score given should be an integer, no decimals|
where MZ i is the average of Z of sample i; cp k is the frequency of the score of sample i is k or more than k;
f (1 − cp k ) is the accumulation of normal function distribution frequency (one side quantile value of normal distribution); m is the evaluation grade, here m is from 1 to 10; SBE i is the SBE value of sample i; and BMMZ is the mean of Z of the benchmark group.
3 Results and analysis
3.1 Three attributes (HSB) of color
3.1.1 Characteristics of hue
The hue was arranged from red (H1) to magenta (H24) on the hue ring (Figure 4). It could be seen that there were 13 hues in total, which were red-purple, purple-toned red, rose, magenta, red, orange-red, orange-yellow, yellow, yellow-green, leaf-green, blue-green, lime green, and green-toned blue. Similar hues were divided into one hue family and finally got six color systems, namely red (23%), orange (24%), yellow (13%), yellow-green (34%), blue-green (5%), and fuchsia (3%). It was found that most of the colors on the ring were warm hue (red, orange, and yellow), and red was the main color. A few of them were neutral and cool hue, and there was no blue, cyan, and purple on the hue ring.
3.1.2 Characteristics of saturation and brightness
The color saturation of species was from 3 to 62% (Figure 5), which was distributed in three levels of low saturation (S1), medium saturation (S2), and medium–high saturation (S3). There were 34 species belonging to S3, which contained the most species of trees, followed by S2. The saturations of most species with a yellow-green color of leaf were concentrated at the medium–high saturation level. Plants with blue-green color leaves had generally low saturation, among which Ophiopogon planiscapus ‘Nigrescens’ had the lowest saturation, i.e., 3%. The saturation of yellow-green family had the greatest change in saturation (the difference was 62%), while that of the purple-red family had the smallest (7%).
The brightness of colors ranged from 15 to 95% (Figure 6), with four stages from low brightness (B1) to high brightness (B4). There were 24 species in the high brightness area, which has the most diversity, and followed by the medium brightness (B2) stage, with 23 species. The brightness of yellow-green series species belonged to B3 and B, while those of the red, orange, and yellow series were mainly distributed in B1 and B2. The brightness difference of each color system was from 24 to 74%, of which the yellow color system has the largest difference of 74%, and the purplish red system has the smallest difference of 24%. The orange (50%), yellow-green (39%), and red (32%) color system were moderate. It was probably because there were more species in the warm-toned system than the cold-toned system, which led to this difference.
3.2 SBE of colored-leaf plants
3.2.1 Evaluation results
A total of 72 people participated in this study as evaluators and produced 72 evaluation forms. Evaluation forms with no difference in scoring for all pictures were defined as invalid forms and eliminated. Finally, there were 65 valid tables left for further analysis.
The ranking of SBE value of 80 plants was shown in Table 3. It could be seen that the plant with the highest score was Oxalis triangularis ‘Purpurea’ (66.65), which has rare hue and peculiar leaf shape. Among the top ten, there were six species that belonged to the yellow-green, accounting for 60%. There were six plants with scores lower than the benchmark value (0.00), and the lowest was Rosmarinus officinalis with a score −7.06.
It could be seen from Table 3 that plants with higher score of SBE tended to have higher color brightness, while plants with lower scores had lower brightness. Therefore, correlation analysis between beauty degrees and brightness should be carried out to find the relationship between them.
3.3 Correlation between the SBE scores and the color attributes
Results of the simple correlation analysis (Table 4) showed that there was a very significant correlation between color brightness and SBE values (P < 0.01), and a significant correlation between saturation of color and SBE values (P < 0.05). But there was no correlation between the hue and the SBE scores.
|Color attributes||Pearson correlation coefficient||Sig. (two-tailed)|
|Hue (x 1)||0.171||0.064|
|Saturation (x 2)||0.201*||0.037|
|Brightness (x 3)||0.534**||0.000|
*Pearson correlation coefficient significant at the 0.05 level (2-tailed), **Pearson correlation coefficient significant at the 0.01 level (two-tailed), main color system (red, orange, yellow, yellow-green, blue-green, and fuchsia).
It could be found in Figure 7 that the hue of plants whose SBE value was greater than the mean value (23.37) mainly belonged to the yellow-green family, and their saturation was concentrated in the middle–high level (26–75%) (Figure 8), and brightness was in the middle–high level (51–100%) (Figure 9). There were 38 species of plants with SBE above the average, of which 31 plants have medium high saturation and 33 species belonged to medium high brightness. This showed that people prefer plants with high brightness and saturation, because bright colors could bring strong sensory and visual stimulation.
3.4 A model of SBE based on color attributes of colored-leaf plants
In order to find out the impact of plant color elements on the beauty degree (SBE), the regression method was used to select the color factors that have a significant impact on the beauty degree. The SBE was taken as the dependent variable, and the three attributes of color were taken as the independent variables. Finally, main factors affecting SBE were selected, and a model was established by multiple linear regression.
By calculations, it was conducted that the brightness (x 3) and hue (x 1) of colors were selected as the main impact indicators that affect the SBE value. The model was as follows:
The variance analysis of the model was extremely significant and the model fitted well, which meant that the model was valid and there was a certain linear relationship between color attributes and the beauty value of colored-leaf plants.
In summary, the brightness of color was the main factor affecting the SBE value of colored leaf plants, followed by hue. Therefore, when colored-leaf plants were used in green space, the most important thing should be considered was the brightness of leaf’s color and then the hue.
4 Conclusion and discussion
The leaf colors of 80 plants contained 13 hues in total, which belonged six color systems. The number of species contained in different color systems, from most to least was yellow–green series (27 species) > orange series (19 species) > red series (18 species) > yellow series (10 species) > blue-green (four species) > purple (two species). There were fewer blue series and purple series species. The brightness and saturation of color were mainly distributed in the middle (26–50%) and middle and high (51–75%) levels, and those of the yellow-green system belonged to the medium and high levels and were significantly higher than other color systems. There has been a significant positive correlation between the brightness and saturation of color and the value of SBE, and an SBE evaluation model was established based on the color attributes.
The color of colored-leaf plants was researched in this article, and an SBE evaluation model was established based on the color attributes. using computer technology, to quantify plant color so as to solve the problem that landscape planners and decision-makers could not define color in landscape design . In previous color studies, RGB color model was usually used to record color, which made it difficult to directly apply the research results in practice. Therefore, in this study, HSB color model was used to quantify color, which is based on the color attributes and conforms to human visual characteristics in color quantification. HSB value and the public’s preference for plant color were used to build a model, which could be directly used in landscape design. Results showed that color brightness was an important factor that affected the value of SBE, which was consistent with the results of other studies [27,28]. According to the evaluation results, people liked plants with high brightness and saturation of color more, such as yellow and green leaf color with higher color brightness and SBE value. When we plan to create a plant color landscape, we can first obtain the HSB value of a plant and input it into the evaluation model to get the SBE value of the plant. The larger the SBE value is, the higher the preference of the plant is. In the main nodes of landscape, plants with medium, medium–high hue, medium–high brightness, or high SBE should be selected for planting.
The color of plants is affected by factors such as temperature, humidity, and sunshine, and the colors of the same plants in different environments are also slightly difference [29,30]. The leaf color of Loropetalum chinense var. rubrum Yieh and Physocarpus opulifolius ‘Diabolo’ in full light is brighter and evener than that in the backlight area [31,32]. Therefore, when collecting samples, it is necessary to detect the ambient brightness to ensure that it is conducted under full light conditions. We tried to take picture under relatively consistent environmental conditions and used a scanner to reduce the error caused by the environment. In addition, because the color characteristics of the same species often differ in different regions or different growth periods, and the samples of plants in this article were only the species commonly used in the current survey area, which will lead to differences in leaf attributes to a certain extent. In the next research, it is necessary to expand the scope of investigation to explore rules of the color changes and leaf color differences of colored-leaf plants from the perspective of Chromatics, so as to provide more theoretical reference for the configuration of colored-leaf species and the construction of the model between color and beauty.
This work is financially supported by the National Natural Science Foundation of China, Grant/Award Number: 31570703, the supplementary funded project of Guizhou province: Guizhou Science and Technology Plan Project (Qiankehe platform talent  No. 5788).
Funding information: The study is supported by the National Natural Science Foundation of China and the supplementary funded project of Guizhou province: Guizhou Science and Technology Plan Project. None of these organizations influenced the study design, the collection, analysis, and interpretation of data, the writing of the report, or the decision to submit the manuscript for publication.
Author contributions: Dr. X.R.W. have full access to all the data in the study and take responsibility for the integrity and the accuracy of the data. Y.J.: acquisition, analysis, or interpretation of data and drafting of the manuscript. X.R.W. and Y.Z.: critical revision of the manuscript for important intellectual content, administrative, technical, or material support, and supervision.
Conflict of interest: No conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication. I would like to declare on behalf of my co-authors that the work described was original research that has not been published previously and not under consideration for publication elsewhere, in whole or in part. All the authors listed have approved the manuscript that is enclosed.
Data availability statement: The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.
|Serial No||Plant species||Family||Genus||growth form||Crown width (m)||height (m)||HSB value|
|1||Prunus virginiana||Rosaceae||Amygdalus||Deciduous trees||3.0–5.0||4.0–5.0||0–33–23|
|2||Prunus cerasifera 'Atropurpurea'||Rosaceae||Prunus||Deciduous trees||3.0–5.0||4.0–5.0||0–32–22|
|3||Cercis canadensis 'ForestPansy'||Euphorbiaceae||Cercis||Deciduous trees||4.0–5.0||4.0–5.0||0–23–17|
|4||Acer platanoides 'Crimson king'||Aceraceae||Acer||Deciduous trees||4.0–5.0||5.0–6.0||0–16–15|
|5||Prunus cerasifera 'Newpiortii'||Rosaceae||Prunus||Deciduous trees||3.0–5.0||4.0–5.0||5–61–36|
|6||Acer rubrum||Aceraceae||Acer||Deciduous trees||3.0–5.0||5.0–6.0||6–60–27|
|7||Acer palmatum 'Europe'||Aceraceae||Acer||Deciduous trees||1.5–2.5||2.0–3.0||9–30–26|
|8||Acer platanoides||Aceraceae||Acer||Deciduous trees||3.0–5.0||5.0–6.0||9–37–20|
|9||Pteris cretica||Aceraceae||Acer||Deciduous trees||3.0–5.0||4.0–5.0||17–23–24|
|10||Quercus rubra||Rosaceae||Prunus||Deciduous trees||1.5–2.5||2.0–3.0||20–50–24|
|11||Hibiscus syriacus 'Argenteovariegatus'||Rosaceae||Prunus||Deciduous trees||4.0–5.0||5.0–6.0||23–33–19|
|12||Prunus serrulata 'Royal Burgundy'||Rosaceae||Cerasus||Deciduous trees||4.0–5.0||5.0–6.0||26–28–20|
|13||Malus 'Royalty'||Rosaceae||Malus||Deciduous trees||4.0–5.0||5.0–6.0||28–31–21|
|14||Quercus rubra||Fagaceae||Quercus||Deciduous trees||4.0–5.0||5.0–7.0||29–72–40|
|15||Amygdalus persica 'Atropurpurea Plena'||Rosaceae||Padus||Deciduous trees||2.5–3.5||3.0–4.0||32–42–24|
|16||Acer × freemanii 'Sienna Glen'||Aceraceae||Acer||Deciduous trees||3.0–5.0||5.0–6.0||43–63–40|
|17||Ulmus pumila 'Golden Leaves'||Ulmaceae||Ulmus||Deciduous trees||3.0–5.0||5.0–6.0||52–55–71|
|18||Ulmus pumila 'Golden Leaves'||Ulmaceae||Ulmus||Deciduous trees||2.0–4.0||2.0–3.0||55–52–80|
|19||Amygdalus persica 'Atropurpurea'||Rosaceae||Prunus||Deciduous trees||1.5–2.5||2.0–3.0||58–40–26|
|20||Lagerstrormia arborescens 'Red leaf'||Chelidaceae||Lagerstroemia||Deciduous trees||1.5–2.5||2.0–3.0||58–47–27|
|21||Acer palmatum 'Orange Dream'||Aceraceae||Acer||Deciduous trees||1.5–2.5||2.0–3.0||58–72–65|
|22||Albizia julibrissin 'Purpurea'||Fabaceae||Albizia||Deciduous trees||2.0–4.0||2.0–3.0||60–51–21|
|23||Cercis chinensis||Fabaceae||Cercis||Deciduous trees||4.0–5.0||5.0–6.0||61–55–79|
|24||Ulmus pumila 'Jinye'||Ulmaceae||Ulmus||Deciduous trees||4.0–5.0||5.0–7.0||61–56–70|
|25||Robinia pseudoacacia 'Aurea'||Fabaceae||Gleditsia||Deciduous trees||4.0–5.0||5.0–7.0||62–60–77|
|26||Ginkgo biloba 'Aurea'||Ginkgoaceae||Ginkgo||Deciduous trees||4.0–5.0||5.0–7.0||63–47–74|
|27||Metasequoia glyptostroboides 'Gold Rush'||Taxodiaceae||Metasequoia||Deciduous trees||4.0–5.0||10.0–13.0||65–46–66|
|28||Fraxinus chinensis 'Aurea'||Oleaceae||Fraxinus||Deciduous trees||3.0–5.0||5.0–7.0||65–46–69|
|29||Sophora japonica 'jinye'||Fabaceae||Styphnolobium||Deciduous trees||4.0–5.0||5.0–7.0||66–55–75|
|30||Sophora japonica 'Jin zhi'||Fabaceae||Styphnolobium||Deciduous trees||2.0–4.0||2.0–3.0||68–44–76|
|31||Acer platanoides 'Princeton Gold'||Aceraceae||Acer||Deciduous trees||1.5–2.5||2.0–3.0||68–61–65|
|32||Acer negundo 'Aurea'||Aceraceae||Acer||Deciduous trees||4.0–5.0||5.0–7.0||70–52–70|
|33||Broussonetia papyrifera 'Jinfeng'||Moraceae||Broussonetia||Deciduous trees||2.0–4.0||2.0–3.0||73–63–70|
|34||Cupressus glabra 'Blue Ice'||Cupressaceae||Cupressus||Evergreen tree||2.0–4.0||2.0–4.0||173–12–54|
|35||Prunus × cistena 'Pissardii'||Rosaceae||Prunus||Deciduous trees||1.5–2.5||2.0–3.0||351–41–20|
|36||Acer palmatum 'Atropurpureum'||Aceraceae||Acer||Deciduous trees||1.5–2.5||2.0–3.0||358–63–47|
|37||Cordyline australis 'Red Star'||Liliaceae||Cordyline||Evergreen shrub||0.4–0.6||1.0–1.5||4–18–33|
|38||Physocarpus opulifolius 'Diabolo'||Rosaceae||Rosaceae||Deciduous shrub||1.2–2.5||1.5–2.5||11–28–24|
|39||Acer palmatum 'Dissectum Nigrum'||Aceraceae||Acer||Deciduous shrub||1.5–2.5||2.0–3.0||18–36–29|
|40||Pteris cretica||Berberidaceae||Berberis||Deciduous shrub||0.3–0.5||0.4–0.6||21–31–48|
|41||Quercus rubra||Hamamelidaceae||Loropetalum||Evergreen shrub||0.4–0.6||1.0–1.5||21–18–37|
|42||Sedum lineare||Rosaceae||Spiraea||Deciduous shrub||0.2–0.4||0.3–0.6||23–57–46|
|43||Sambucus nigra 'black lace'||Loniceraceae||Sambucus||Evergreen shrub||1.5–2.5||2.0–3.0||42–28–18|
|44||Cotinus americana 'Royal Purple'||Aceraceae||Acer||Deciduous shrub||1.5–2.5||2.0–3.0||43–26–21|
|45||Melaleuca bracteata||Myrtaceae||Melaleuca||Evergreen shrub||2.5–3.5||3.0–4.5||53–50–76|
|46||Buxus sempervirens 'Aurea'||bupleurumceae||Buxus||Evergreen shrub||0.4–0.6||0.6–1.0||53–42–95|
|47||Ligustrum sinense 'Jin ye'||Oleaceae||Ligustrum||Deciduous shrub||0.4–0.6||0.6–1.0||59–63–77|
|48||Forsythia koreana 'Sun Gold'||Oleaceae||Forsythia||Deciduous shrub||0.3–0.5||0.6–1.0||62–65–72|
|49||Physocarpus opulifolius 'Luteus'||Rosaceae||Physocarpus||Deciduous shrub||0.4–0.6||0.6–1.0||63–64–60|
|50||Platycladus orientalis 'Beverleyensis'||Cupressaceae||Platycladus||Evergreen shrub||2.5–3.5||3.0–4.5||69–52–73|
|51||Wisteria sinensis||Fabaceae||Wisteria||Deciduous vine||2.5–3.5||3.0–4.5||70–54–77|
|52||Ligustrum × vicaryi||Oleaceae||Ligustrum||Evergreen shrub||0.4–0.6||0.6–1.0||70–70–61|
|53||Caryopteris×clandonensis 'Worcester Gold'||Verbenaceae||Caryopteris||Deciduous shrub||0.4–0.6||0.6–1.0||71–62–66|
|54||Platycladus orientalis 'Semperaurescens'||Cupressaceae||Platycladus||Evergreen shrub||1.2–2.0||1.5–2.5||74–52–70|
|55||Platycladus orientalis 'Aurea Nana'||Cupressaceae||Platycladus||Evergreen shrub||1.0–1.5||1.2–2.5||77–54–71|
|56||Teucrium fruticans||Labiatae||Teucrium||Evergreen shrub||1.0–1.5||1.2–2.5||80–19–42|
|57||Cupressus macrocarpa 'Goldcrest'||Cupressaceae||Sabina||Evergreen shrub||1.2–2.0||1.5–2.5||81–53–73|
|58||Picea pungens||Pinaceae||Picea||Evergreen shrub||2.0–4.0||2.0–4.0||142–17–58|
|59||Pennisetum setaceum 'Rubrum'||Rosaceae||Pennisetum||Perennial herb||0.15–0.2||0.25–0.35||5–56–28|
|60||Heuchera micrantha||Saxifragaceae||Heuchera||Perennial herb||0.15–0.2||0.25–0.35||9–41–39|
|61||Euonymus japonicus 'Marieke'||Poaceae||Pennisetum||Perennial herb||0.15–0.2||0.25–0.35||24–40–24|
|62||Gaura lindheimeri 'Crimson Bunny'||Salicaceae||Gaura||Perennial herb||0.15–0.2||0.25–0.35||29–39–27|
|63||Canna warscewiezii||Canniaceae||Canna||Perennial herb||0.15–0.2||0.25–0.35||29–46–28|
|64||Cyperacea Carex||Cyperaceae||Carex||Perennial herb||0.15–0.2||0.25–0.35||34–58–59|
|65||Stipa lessingiana||Poaceae||Stipa||Perennial herb||0.15–0.2||0.25–0.35||42–37–68|
|66||Lysimachia nummularia 'Aurea'||Primulaceae||Lysimachia||Perennial herb||0.15–0.2||0.25–0.35||60–64–75|
|67||Sedum lineare||Crassulaceae||Sedum||Perennial herb||0.15–0.2||0.25–0.35||68–53–82|
|68||Coleus blumei||Labiatae||Coleus||Annual herb||0.15–0.2||0.25–0.35||70–68–80|
|69||Ipomoea batatas 'Tainon'||Convolvulaceae||Ipomoea||Perennial herb||0.15–0.2||0.25–0.35||71–61–76|
|70||Seneccio cineraria 'Silver Dust'||Compositae||Senecio||Perennial herb||0.15–0.2||0.25–0.35||75–8–79|
|71||Rosmarinus officinalis||Labiatae||Rosmarinus||Perennial herb||0.15–0.2||0.25–0.35||84–24–41|
|72||Dianthus gratianopolitanus||Caryophyllaceae||Dianthus||Perennial herb||0.15–0.2||0.25–0.35||86–28–53|
|73||Festuca glauca||Poaceae||Festuca||Perennial herb||0.15–0.2||0.25–0.35||165–11–57|
|74||Ophiopogon planiscapus 'Nigrescens'||Liliaceae||Ophiopogon||Perennial herb||0.15–0.2||0.25–0.35||180–3–15|
|75||Oxalis triangularis 'Purpurea'||Oxaliaceae||Oxalis||Perennial herb||0.15–0.2||0.25–0.35||290–39–52|
|76||Ajuga ciliata||Labiatae||Scutellaria||Perennial herb||0.15–0.2||0.25–0.35||324–46–28|
|77||Setcreasea purpurea||Commelinaceae||Tradescantia||Perennial herb||0.15–0.2||0.25–0.35||338–19–23|
|78||Begonia semperflorens||Begoniaceae||Begonia||Perennial herb||0.15–0.2||0.25–0.35||348–45–25|
|79||Iresine herbstii||Amaranthaceae||Iresine||Perennial herb||0.15–0.2||0.25–0.35||350–57–30|
|80||Imperata cylindrical 'Rubra'||Poaceae||Imperata||Perennial herb||0.15–0.2||0.25–0.35||352–45–27|
 Ye XQ, Wang L, Guan HW, Zhou S, Yu R, Shen XD. Landscape construction and analysis of autumn color plants in Northwest China—A case study of the autumn garden in Yinchuan botanical garden. J Chin Landsc Archit. 2018;34(S1):94–7.Search in Google Scholar
 Tang C. Effects of pigment indicators on the color expression of leaves of the colored-leaf garden plants. Agric Sci Technol. 2014;15(11):1928.Search in Google Scholar
 Badami AA. Management of the image of the city in urban planning: experimental methodologies in the colour plan of the Egadi Islands. Urban Des Int. 2022;1–6. 10.1057/s41289-022-00200-1.Search in Google Scholar
 Zhang L, Cao DM, Hu JJ. Advance of the regulation mechanism of leaf color formation in plants. J Plant Genet Resour. 2021;22(02):293–303.Search in Google Scholar
 Chen JJ, Liu BG, Rui L, Liu Y, Li YH. Research on plant configuration based on quantitative analysis of plant community color composition. J Henan Agric Univ. 2019;53(4):550–6.Search in Google Scholar
 Yang L, Wang XR, Li YQ, Yang J, Han S. Quantitative analysis of plant community landscape color in Guanshan Lake Park of Guiyang City. J Southwest Univ. 2021;41(3):1–10.Search in Google Scholar
 Li H, Zhuo L. The classification of colorful-leaf tree species and the application in landscape architecture. J Northeast Agric Univ (Engl Ed). 2008;1:67–9.Search in Google Scholar
 Shen XH. Landscape dendrology. Chongqing, China: Chongqing University Press; 2013.Search in Google Scholar
 Zhang YK, Wang XR, Yang T, Zhuo L, Bai C. Change characteristics of new leaf color attributes of common garden plants in spring. J Northwest Univ. 2019;34(6):208–13.Search in Google Scholar
 Feng SN, Yue H. Quantitative study on seasonal color of non- agricultural habitats landscape in cold regions agricultural sightseeing garden. J Northeast Agric Univ (Engl Ed). 2018;49(7):27–37.Search in Google Scholar
 Li C, Shen S, Ding L. Evaluation of the winter landscape of the plant community of urban park green spaces based on the scenic beauty estimation method in Yangzhou, China. PLoS One. 2020;15(10):e0239849.10.1371/journal.pone.0239849Search in Google Scholar PubMed PubMed Central
 Fujimoto T, Otani M, Nakano M. Production of colored foliage phenotypes in Kalanchoe blossfeldiana by ectopic expression of R2R3 MYB genes. J Plant Biochem Biot. 2022;31:1–8.10.1007/s13562-021-00760-3Search in Google Scholar
 He M, Wang H, Hu Y, Zhao P, He S, Mei Y. Research status, development and utilization of color-leaf trees. Acta Agric Univ Jiangxiensis. 2018;40(6):1134–44.Search in Google Scholar
 Shen SG, Li CL, Su L. Research progress in plant colorscape based on visual perception. World For Res. 2021;34(1):6.Search in Google Scholar
 Macfarlane C, Ogden GN. Automated estimation of foliage cover in forest understorey from digital nadir images. Methods Ecol Evol. 2012;3(2):405–15.10.1111/j.2041-210X.2011.00151.xSearch in Google Scholar
 Kendal D, Hauser CE, Garrard GE, Jellinek S, Giljohann KM, Moore JL. Quantifying plant colour and colour difference as perceived by humans using digital images. PLoS One. 2013;8(8):e72296.10.1371/journal.pone.0072296Search in Google Scholar PubMed PubMed Central
 Jia Y, Ito T, Lee JJ, Sugisaka M. Histogram matching based on gaussian distribution on the HSB color system. International Conference on Artificial Life and Robotics (ICAROB); 2016. p. 36–9.10.5954/ICAROB.2016.OS11-2Search in Google Scholar
 Frank S, Fürst C, Koschke L, Witt A, Makeschin F. Assessment of landscape aesthetics—Validation of a landscape metrics-based assessment by visual estimation of the scenic beauty. Ecol Indic. 2013;32:222–31.10.1016/j.ecolind.2013.03.026Search in Google Scholar
 Li H, Shi K, Wang Y, Li Y, Feng Y. Research on scenic beauty estimation of plant landscape on the roof on SBE method. Arab J Geosci. 2021;14(10):1–8.10.1007/s12517-021-07225-wSearch in Google Scholar
 Qiangmin ZH, Chuanwei HU, Bing SU, Meina GA, Xue GA. Aesthetic evaluation and application of ornamental plant landscape. J Landscape Res. 2017;9(5):103–5.Search in Google Scholar
 Daniel TC, Boster RS. Measuring Landscape esthetics: the scenic beauty estimation method. US: USDA Forest Service Research Paper RMRS; 1976.Search in Google Scholar
 Ma BQ, Xu CY, Cui Y. Effects of color composition in badaling forests on autumn landscape quality. Northwest Univ. 2018;33(06):258–64.Search in Google Scholar
 Yu Y, Yao CC, Zhang J, Qi JZ. Scenic beauty estimation on the common green plants monomers of two cities in Jilin. Beihua Univ Nat Sci. 2017;18(4):521–7.Search in Google Scholar
 Dong JW, Liao YM, Xu XS, Zhang Z, Wang T. Evaluation on the beauty of ornamental plants in autumn. J Northwest Univ. 2010;38(3):42–6.Search in Google Scholar
 Chiang C, Bnkestad D, Hoch G. Reaching natural growth: The significance of light and temperature fluctuations in plant performance in indoor growth facilities. Plants (Basel, Switz). 2020;9(10):1312.10.3390/plants9101312Search in Google Scholar PubMed PubMed Central
 Dai XH, Zhou LY, Hou LQ. Relationship between color variation and pigments in Loropetalum chinense var. rubrum leaves under different light conditions. J Northwest Univ. 2015;43(3):62–5.Search in Google Scholar
 Zhang H, Zhong H, Wang J, Sui X, Xu N. Adaptive changes in chlorophyll content and photosynthetic features to low light in Physocarpus amurensis Maxim and Physocarpus opulifolius “Diabolo”. Peer J. 2016;4(3):e2125.10.7717/peerj.2125Search in Google Scholar PubMed PubMed Central
© 2022 the author(s), published by De Gruyter
This work is licensed under the Creative Commons Attribution 4.0 International License.