Urban fringe areas are social-ecological vulnerable areas at the interface of urban and rural areas, and their habitats are deeply influenced by urban expansion and rural urbanization. Assessing the plural values of habitat services in urban fringe areas is crucial for environmental protection and sustainable development. Most of the current studies on habitat services, however, either focus on biophysical processes and ecological quality or consider social perceptions, and studies integrating both remain rare in urban fringe areas. This could be detrimental to the formulation of conservation and management strategies with broad acceptance. Taking Qingpu District in Shanghai as an example, we collected its physical geographic and land use/land cover data, and obtained residents' perceptions through a questionnaire survey including participatory mapping (n=223). We used the InVEST model and SolVES model to assess habitat quality and habitat perceptions, respectively, from the perspective of social-ecological analysis. Furthermore, we analyzed the coupling coordination relationship between them using the coupling coordination degree model, and investigated the influencing factors of the variation of the coupling coordination using Geodetector. We finally developed habitat services conservation and management zoning based on the above. Our results showed that: (1) the habitat degradation of the study area was at a relatively weak degradation level, with a mean value of 0.224. Habitat quality was at a relatively low level, with a mean value of 0.340, and the spatial pattern was high in the west and low in the east. In terms of social value, an average of 1.37 points of habitat perceptions were mapped per person. These points were distributed mostly in public green areas, water areas, and several villages with good greenery. The spatial distribution of habitat perceptions was more even, with moderate and high values distributed in striped patterns along the water areas. (2) The coupling and coordination of habitat quality and habitat perceptions were in mild incoordination, with a mean value of 0.428. There were more units in mild incoordination, with a percentage of 35.90%. The spatial pattern of coupling and coordination was also high in the west and low in the east. (3) Socio-economic factors (e.g.,GDP and road density) were the main influencing factors of the spatial heterogeneity of coupling and coordination, and they mostly interacted with landscape pattern factors (e.g., patch richness and Shannon diversity) to enhance the explanation level of spatial variation. No significant influences were found for landscape connectivity on the spatial heterogeneity of coupling and coordination. (4) Based on the high-low matching level as well as the coupling and coordination relationship between habitat quality and perceptions, the study area could be divided into four major categories and twelve subcategories. Targeted conservation and management strategies were proposed for each area based on local conditions. Our study integrated the complementary intrinsic value and relational value, and the results could provide a basis for comprehensive habitat conservation that promotes coordinated development of ecology and society as well as harmonious coexistence between people and nature.