In order to improve the effect of garden landscape design in the residential area of sponge city, this paper combines digital technology to carry out the garden landscape design of the residential area of sponge city, and conducts the research on it through the intelligent system. Moreover, this paper studies the digital color processing algorithm and proposes a digital technology-based color processing algorithm for the garden landscape design of the residential area of the sponge city. In addition, this paper builds a garden landscape design system for sponge city residential areas on the basis of algorithms. Through the simulation test research, we can see that the garden landscape design of sponge city residential area based on the digital technology proposed in this paper can meet the needs of the sponge city garden landscape design.

1. Introduction

Sponge city is a new generation of urban stormwater management concept, which means that the city has good “elasticity” in adapting to environmental changes and responding to natural disasters caused by rainwater, and can also be called a “water elastic city.” It is likened to a city like a sponge, which can absorb, store, infiltrate, and purify rainwater on the spot when there is rainfall, replenish groundwater, and regulate the water cycle. However, in the event of drought and water shortage, it has the conditions to release the stored water and make use of it. The international term is “Low-Impact Development Stormwater System Construction.” Its essence is to change the rainwater through pipelines, pumping stations, and other gray facilities to quickly discharge the terminal concentrated mode to the slow discharge and slow-release source dispersion mode through rain gardens, green roofs, wetlands, and other green “sponges.” The sponge city advocated is to organize and link the rivers, lakes, and groundwater systems in the city through ecological infrastructure and engineering technology to form a cooperating whole. These include a variety of green “sponges” such as rain gardens, green roofs, ecological wetlands, and forests.

The proposal of the sponge city theory marks a new understanding of the urban rainwater problem, and its adoption in official documents means that the ecological rainwater management ideas and technologies have moved from academia to the management level and have been effectively promoted in practice. The residential green space occupies an important position in the urban green space, and the green area is relatively large and the environmental benefits are good, which undoubtedly provides good conditions for the “sponge city” and urban ecological rainwater management and the construction of rainwater gardens.

In the continuous expansion of the city, high-rise residential buildings are rising everywhere, and the proportion of impermeable roads is increasing, resulting in rain disasters in low-lying areas every rainy day. The idea of a sponge city is to lay rainwater pipes in the traditional way while adopting development technology with less impact so that hydrology and ecology cannot be affected, and in the event of natural disasters, it has the “elasticity” similar to a sponge. In the construction of a sponge city, it needs to be controlled at the source, in the middle and in the later stage with the help of less influential development technical solutions. Source control measures include green roofs, shallow ditch for vegetation, seepage wells, early rainwater abandonment, etc.; mid-term control measures include rainwater sedimentation ponds, permeable pavement, retention ponds, sunken green spaces, etc.; and late-stage control measures include yes, wet ponds, stabilization ponds, infiltration ponds, constructed wetlands, rain gardens, etc. The requirements of modern people for the living environment are not only limited to the grade and quality of the house, but the requirements for the living environment are getting higher and higher. Garden landscape is an important part of the urban living environment and plays an important role in improving the living environment and quality of life in residential areas. Therefore, the quality of residential landscape design also determines the spatial quality of the entire residential project to a large extent. Landscape plays a vital role in the success or failure of a project.

In this paper, digital technology is used to design the garden landscape of the residential area of the sponge city, and the research of the garden landscape design of the residential area of the sponge city is carried out through the intelligent system, so as to improve the effectiveness of the garden landscape design of the residential area of sponge city.

Literature [1] studies the recycling of urban rainwater adopts the sunken optimization design method and transforms the urban green space square, which can not only prevent the long-term erosion of rainwater but also prevent part of the green space from subsidence, allowing rainwater to penetrate into the city. In the soil, the natural cycle is completed. Literature [2] carried out in-depth research on the reconstruction plan, the purpose of which is to effectively solve and utilize the natural precipitation problem based on the concept of ecological sponge city and the method of “ecological drainage + water drainage.” Literature [3] conducted a detailed investigation and research on wetland parks, which effectively solved the problem of urban waterlogging. The foundation is the rainwater safety pattern. As a green sponge complex, the function of the wetland park is mainly to carry urban rainwater. The urban rainwater is collected into the wetlands, and the wetlands are used for digestion. The development model of “water-collecting urban areas-water-collecting wetlands” has been formed [4]. The multilevel sponge terrain is created at a low cost by the fill and dig technology, and the multilevel wetland system is constructed to form a more favorable topography for the system [5]. The construction of a multilevel and multifunctional wetland system can realize the functions of gathering, purifying, and replenishing groundwater. In addition to realizing ecological circulation, it can also play a role in soil and biological purification. The purified rainwater flows into low-lying areas to replenish groundwater resources [6].

Literature [7] introduces the concept and function of planting shallow grass trenches in detail and expounds on its design. Literature [8] explores the rainwater utilization landscape of different sites through the research on the rainwater block project and summarizes the landscape rainwater facilities such as road tooth expansion pool, rainwater lawn planting, rainwater garden planting belt, and grass planting ditch. Literature [9] explored different aspects of rainwater utilization and the concept of green streets and proposed different gardening elements to improve the rainwater treatment capacity. Literature [10] studies international stormwater management from the perspective of landscape optimization and proposes the selection and collocation method of plant species on-site under the premise of stormwater management. Literature [11] discusses rainwater management from the perspective of landscape ecology, proposes the concept of green rainwater facilities, and explains the security pattern of landscape ecology. Literature [12] studies the aesthetics and safety aspects of rainwater landscape construction and proposes optimization measures and optimization design procedures for flood management facilities.

Literature [13] conducted a comprehensive study on the natural hydrological conditions of the city and pointed out the related problems of the overall planning of the sponge city. Only with the help of physical or biological technology, the natural ecology can be fully utilized to build perfect urban drainage, leakage prevention, roads, green space, and other systems; reduce the urban land area as much as possible; and restore the ecological appearance of the city. Literature [14] deeply studied the urban low-impact development of rainwater systems and proposed that to build a sponge city, the impact on the water circulation system must be reduced, and the green space, squares, wetlands, etc., in the city must be fully utilized. Literature [15] deeply analyzed the problems faced by the construction of sponge cities, and put forward the following suggestions: first, we should take the government the leading role to establish a perfect water cycle treatment system; secondly, we should strengthen the management of the approval process, and at the same time attract capital from all aspects. attention, broaden the source of funds, then establish an incentive mechanism or business model to encourage and promote the construction of sponge cities, and finally call on the whole society to pay attention to and participate in urbanized water environment governance, which can attract international cooperation to improve governance and construction. le Polain de Waroux et al. [16] pointed out that only by breaking the limitations of traditional concepts and realizing innovation in urban construction can we build a new ecological civilization city, promote the development of sponge cities, use ecological land, and create an ecological water cycle system. Literature [17] believes that the management of rainwater resources is currently in a disordered state and has not been incorporated into the overall planning of urban water resources. Rainwater resources are not only underutilized but also have become an important factor in urban waterlogging. Rainwater should be considered in planning.

Literature [18] summarizes and classifies stormwater facilities in combination with the concept of low-impact development of sponge cities and evaluates the design effect of Shangyang Avenue through the construction of SWMM, thus confirming that the road of the concept of a sponge city is closer to the hydrological environment than before. Literature [19] takes landscape control of stormwater as a new perspective and summarizes the calculation methods and selection methods related to stormwater design under different stormwater control objectives. Literature [20] explores the design of rainwater gardens, which includes rainwater design. The types and functions of gardens, the design and calculation methods of rain gardens in different spatial states, and the selection of relevant green plant species. It also analyzes the influencing factors of the growth environment of green plants and explores the matching methods and methods of green plants and their later maintenance measures.

3. Digital Garden Landscape Color Processing Algorithm

The CIEXYZ color space was born to solve the problem of negative values in the CIE1931RGB color space. CIERGB needs to use negative values when matching the color of visible light, which is not only inconvenient to use but also difficult to understand. The three primary colors of X, Y, and Z selected by the CIE do not correspond to visible colors. They are the three primary colors in my imagination. The specific values are derived from the CIE1931RGB system. Among them, all the X, Y, and Z values in CIEXYZ are positive numbers, and the human eye's response to brightness is represented by the Y value. The conversion relationship between CIEXYZ and CIERGB is shown in the following formula:

After the transformation coefficient is normalized, it is shown in the following formula:

It is worth noting that, in practical applications, different color devices have different conversion relationships due to different color characteristics, and these relationships are recorded in the color profile file.

The CIELAB color space is a nonlinear transformation space of the CIEXYZ space, which was developed to solve the inhomogeneity of the chromaticity space in the past. The CIELAB uses L, a, and b , the three different coordinate axes to describe any color in nature, where L represents lightness, the bottom is black, and the top is white. a means red/green, +a means red, −a means green, b means yellow/blue, +b means yellow, and −b means blue. Studies have shown that the prediction of hue in CIELAB space is not accurate, especially in the prediction of blue hue, which will not be described in detail here. The three-dimensional structure of the CIELAB color space is shown in Figure 1:

The conversion formula between the CIELAB color space and the CIEXYZ color space is as follows:

In the following formula (3), XYZ is the sample color tristimulus value, is the reference white tristimulus value under the standard light source, and generally the tristimulus value of a D50 light source or D65 light source is used.

The PCS under the ICC specification is responsible for two tasks. One is the chromaticity purpose conversion, and its PCS value is the original chromaticity value. Another task is perceptual purpose transfer, and its PCS value is the perceptual property of the color appearance of the replicated image in the reference viewing environment. Figure 2 shows the role of PCS color space in color management.

PCS link space is generally CIEXYZ and CIELAB. This is because there is a simple linear conversion between CIEXYZ and most of the international standardized RGB spaces such as PAL RGB, NTSC RGB, and sRGB. However, the CIEXYZ space is a space with nonuniform color vision, and the mapping relationship with the printer CMYK space cannot be expressed by analytical expressions. In order to make up for this deficiency, the ICC stipulates that CIEXYZ is used to define the chromaticity value of the original and then converts it into the CIELAB color space to describe the human eye’s perception properties under observation conditions. The role of the PCS color space under the ICC standard in the color management system is shown in Figure 3.

It should be pointed out that there is inhomogeneity in the hue of the CIELAB color space, conversion errors will inevitably occur during color gamut mapping, and the color difference formula based on CIELAB can only calculate color blocks, but not calculate the color difference of complex space images.

CIELAB, CIEXYZ, or CIECAMO2 are all based on simple color blocks to describe color, and none of them can meet the prediction of image color appearance related to visual adaptation and visual spatial and temporal characteristics. However, whether it is a color electronic image or a printed product, it cannot be a single color block, so PCS should be undertaken by an image color appearance model. The image color appearance model iCAM is established on the basis of CIECAM02, which takes into account that the human eye changes its perception of white point with the difference of spatial frequency. Therefore, after using CSF to simulate the human eye filtering process, the maximum stimulus point of the image is used as the reference white. The IPT (Intensity Protan Tritan) color space is both a uniform color space and an opposite color space. Its main advantage is that the prediction of hue is more accurate than in other color spaces in the past, and it can be used in the color reproduction process to improve the reproduction quality of printed matter. iCAM finally transforms the color into the IPT opposite color space, which not only simplifies the transformation method of CIECAMO2 but also predicts the hue more accurately. Figure 4 shows the role relationship of the iCAM color appearance model as the PCS link space in the color management system.

In addition to the above-mentioned PCS color space selection method based on color appearance, there is also a spectrum-based color management and replication technology, as shown in Figure 5. Metamerism is ubiquitous in color reproduction, and if PCS takes the form of spectral reflection, two spectrally matched colors will maintain the same appearance in any lighting environment. In addition, the spectral data, as the original data of color generation, retains the most information and the greatest flexibility, realizes the accurate prediction of color, and can fundamentally eliminate the fundamental defects of traditional printing and color management technology.

The spectrum-based color management method is still in the theoretical research stage, and the specific implementation requires professional equipment and technical personnel. There is still a long way to go before it can be widely used in practice. The main difficulty is that the digital channels of color physical devices cannot be too many, so the spatial conversion between the spectral data of dozens of channels and the digital drive of the device is not an injective relationship. The acquisition of spectral data, how to realize the color gamut description of high-dimensional spectral data, color gamut matching, and color correction, data processing of multispectral images, and how to store, compress, transmit, and the display is the focus and difficulty of research.

The human visual system is a very complex system. When it perceives external color stimuli, the size, structure, shape of the stimulus itself, the external environment, and the state of the observer can all affect it. Therefore, it is impossible to accomplish the purpose of color management only by establishing the tristimulus value model of color.

When two color samples with the same tristimulus value, under different observation environment, background, sample size, sample shape, etc., the human visual perception is different. Moreover, different backgrounds will have different color perceptions, as shown in Figure 6.

In general, the color changes toward the complementary color of the background color; that is, if the background is darkened, the color will be brighter. If the background is redder, the color looks greener; the yellower the background, the bluer the color feels, which is called the color appearance phenomenon. There are many kinds of color appearance phenomena, such as simultaneous contrast, amplification, Hunt effect, Stevens effect, Abney effect, etc., which will not be introduced in detail here. Chromatic adaptation transformation appears to solve the color appearance phenomenon of color perception under different light sources or observation conditions, and its transformation parameters are based on a large number of corresponding color data sets.

The environmental parameters of the CATO2 model are selected, and then the color adaptation transformation can be performed:

Step 1. First, the tristimulus values XYZ are converted to the cone response space, as shown in the following formula:Among them,

Step 2. The algorithm calculates the adaptation factor D according to the environmental parameters, as follows:

Step 3. The algorithm calculates the cone response under the reference white, as follows:Finally, the inverse transform calculates the corresponding color; that is, the tristimulus value after adaptation, as shown in the following formula:The color appearance model includes two parts: color adaptation and color appearance attribute space. First, the input XYZ tristimulus values are transformed into the cone-responsive color space through color adaptation. Then, the color appearance attribute value is calculated according to the adapted color signal. The forward transformation steps are shown in Figure 7:(1)XYZ tristimulus value changes to human eye pyramidal cell response.(2)The algorithm inputs environmental parameters and performs color adaptation transformation after cone response.(3)The algorithm transforms into the color appearance attribute space according to the adaptive cone response.(4)The algorithm simulates the nonlinear compression of the visual system.(5)The algorithm calculates various color appearance attributes.CIELAB space is the simplest color appearance model, and its calculation steps are as follows:
The first step is to enter the corresponding model environment parameters. In the second step, according to the environmental parameters, formula (4) is used to perform color adaptation transformation, and the XYZ values are transformed into the vertebral body response space under the reference conditions. The third step is to perform physiological color space conversion. The cone response space after chromatic adaptation transformation is transformed into a cone response HPE space close to the physiological one, and the transformation relationship used is shown in the following formula: Among them,The fourth step is to simulate the dynamic nonlinear characteristics of the cone response of the human visual system according to the following formula:Finally (the fifth step), according to formula (12)–(16), the correlated color attribute is calculated.(1)“no color response” A.(2)Brightness.(3)Apparent brightness.(4)Temporary amount.(5)Chroma c, visual chroma M, and color saturation s.After the above steps, the color appearance attribute conversion analysis based on the CIECAMO2 model can be realized.
The CIECAMO2 model is adopted in the latest windows color management system, WCS. Specifically, when performing color gamut mapping, the JCH space based on the CIECAMO2 color appearance model is used as the PCS link space. The CIACAM02 model is used as the PCS space for color gamut conversion, and the process is shown in Figure 8.
In order to take into account the influence of color appearance perception and color difference resolution, S-CIELAB first transforms the observed image to the opposite color space when processing the image. At the same time, the corresponding classification is carried out, and then the spatial fuzzy characteristics of the human visual system are simulated, and CSF filtering is performed for each classification. Finally, the filtered color values are inversely transformed into the CIEXYZ and CIELAB spaces in turn. The whole process is shown in Figure 9.
The detailed steps of their conversion include the foollowing:(1)According to formula (17), the algorithm transforms the image color information XYZ into three opposite colors AO1O2 as follows:Among them, the three opposite colors are black-white, red-green, and yellow-blue.(2)The algorithm performs 2-D convolution calculation on each color component, and its convolution kernel is shown in the following formula:Among them,In the above formula, K and are normalization coefficients. Different weights and variable parameter values are obtained through human eye victory experiments.(3)According to formula (20), the algorithm transforms the color information AO1O2 processed by spatial filtering into the CIEXYZ space:(4)The algorithm transforms CIEXYZ to CIELAB. The specific conversion formula is the same as before, and will not be repeated here.The main advantage of IPT space is that the prediction of hue is more accurate than other color spaces, so Fairchild calls it the processing space of the iCAM appearance model. The color properties of IPT space and CIELAB space are the same, I stands for lightness, Р stands for red/green, and T stands for yellow/blue. However, the value range of IPT is different from that of CIELAB. The value of I is (0, 1), and the value of P and T is {−1, 1}. The input data of IPT is the XYZ value of the observer under the D65 light source and the 2° field of view, and the color space LMS is responded to by the cone cells. The transformation relations used are shown in formulas (21)–(23).The processing process of the iCAM model is similar to that of CIECAMO2, and the input parameters and calculation formulas have not changed, including tristimulus values, adaptive white point, adaptive brightness, and environmental factors. Among them, only the definition of the environmental parameters has changed. After the human eye CSF filtering process is used, the maximum value of the image stimulus is used as the reference white point. This change makes the iCAM model independent of specific viewing conditions when dealing with color, which is not possible with other models. Finally, iCAM converts the color to the IPT color space, which not only greatly simplifies the calculation steps but also improves the accuracy of hue prediction.
The following are the calculation steps of the iCAM model:
The first step is to perform a color-adaptive transformation on the original image and the low-pass filtered image. The CATO2 color adaptation transformation model is adopted, and the specific formula is as follows:Step 1: First, the model converts the original image and the low-pass filtered image to the cone response space RGB of the human visual system, as follows:The original image [21].Low-pass filtered image.The model calculates the adaptation factor, and the adaptation brightness and environmental factors are set according to the external environment:The cone responses after color adaptation are as follows:Step 2: The adapted cone response RcGcBc is transformed into the IPT color space, first transformed to the corresponding color XD65Y65Z65 in the D65 reference environment:Then, CIEXYZ transforms L to IPT, which is a little different from the method mentioned above. The adaptive brightness factor F is added to adjust the nonlinearity in the transformation, as shown in the following formula:The model calculates the luminance factor FL:Its transformation to IPT space is shown in the following formula:Step 3: The last step is the color appearance attributes, brightness J, chroma C, hue h, apparent brightness Q, apparent chroma M, and image difference Im as follows:

4. Garden Landscape Design of Sponge City Residential Area Based on Digital Technology

Figure 10 shows the virtual interface structure of the garden landscape design system in the residential area of the sponge city based on digital technology.

According to the collected environmental landscape information, the location processing of the landscape position to be processed is carried out, and the network technology and infinite network nodes are further used to carry out comprehensive virtual auxiliary processing of the environmental landscape design, so as to improve the efficiency of the environmental landscape virtual design. In the process of 3D simulation of environmental landscape, it is necessary to take into account factors such as regional landscape topography information, vegetation planting information, building appearance, and building feature data, and carry out three-dimensional design and rendering of the scene to achieve multiangle and multidirectional construction, editing, and restoration of landscape information. The specific environmental landscape three-dimensional simulation information processing module is shown in Figure 11.

Figure 12 below shows the effect of the garden landscape design in the residential area of the sponge city.

The model proposed in this paper is simulated by Matlab, and the landscape design effect of the sponge city residential garden landscape design system based on digital technology is counted as shown in Table 1.

From the above research, it can be seen that the garden landscape design of sponge city residential area based on digital technology proposed in this paper can meet the needs of sponge city garden landscape design.

5. Conclusion

In the process of urban construction, a large number of hard impermeable pavements are used, so that more and more natural land is replaced by impervious pavement, and the original permeable layer of natural precipitation is cut off, interrupting the original urban water cycle. There are many environmental problems in cities, such as massive loss of rainwater resources, serious runoff pollution, and increased risk of flood disasters. The traditional urban stormwater management adheres to the planning concept of “come quickly and go quickly,” while the municipal drainage system quickly drains the rainwater, which forms an obvious contradiction. On the one hand, due to the excessive exploitation of groundwater, the groundwater level has dropped, resulting in a serious water shortage in the city. On the other hand, urban rainwater is not used reasonably, and a large amount of discharge causes waste and even damages the urban ecological environment. This paper combines digital technology to carry out the garden landscape design of the residential area of the sponge city, and conducts the research on it through the intelligent system. The research results show that the garden landscape design of sponge city residential area based on the digital technology proposed in this paper can meet the needs of sponge city garden landscape design.

Data Availability

The labeled dataset used to support the findings of this study is available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.


The research was supported by the 2017 Youth Project for Humanities and Social Science Research of the Ministry of Education and funded by the results of the “LID Landscape Model Study in Sponge City Construction” (No. 17YJC760077).