Applications of image processing in viticulture: a review

Abstract

The economic importance of the wine industry worldwide is driving the development and application of innovative methods and technologies for monitoring vineyards. The harvest can significantly vary from year to year and also within the due to soil conditions, disease, pests, climate and variation in management practices. Current practices for yield and quality estimation are destructive, expensive, inaccurate and largely subjective. These factors make the production of high quality grapes for wine making challenging. Image analysis has the potential to provide an inexpensive, non- destructive way of capturing precise information about the entire vineyard. This paper presents a review of recent research relevant to the application of digital image analysis to the management of vineyards focusing on the key challenges for in-field, on the vine, ground level image capture and analysis. The applications explored are: Yield Estimation: Historically efforts have primarily been focused on the objective estimation of harvest yield. In order to estimate yield using image processing the grapes must be detected, segmented from the background and the number of clusters counted and the size of the cluster determined and this is typically how most estimation methods, using images, are approached. However for accurate yield forecasting berries within the cluster must be counted and the size of each berry determined. This is difficult to achieve due to the occlusion of berries. Many of the most successful grape detection methods have been developed, independently of the work on yield forecasting, for the automation of management activities such as harvesting and smart spraying rather than for yield estimation. Quality Evaluation: Berry quality can be linked to visual properties such as the size, weight and colour of the berry. These properties have been proven to be indicators for ripeness and level of phenolics, flavonoids and sugar. To date most of the work in this area has focused on the processing of images of individual grapes in the laboratory. In the natural environment approaches must be able to cope with difficult and variable lighting conditions and with factors such as occlusion of the grapes within the cluster and by the canopy. Disease Detection: While some work has been undertaken towards the detection of powdery and downy mildew and botrytis in grapes there is still significant scope for improvement. Detection in the field is difficult as the grapes may be covered by bloom and disease can exhibit different signs and symptoms depending on the grape variety and the stage of development of the disease. Moreover, more than one disease might be present. Current work does not address the issues of more than one disease or identification of the disease at various stages in its development. The most successful work to date has not been on in-field images. Grape Phenology: The phenology of the grape vine is complex. Understanding the phenology of a given plant system is important in determining the ability of a region to produce a crop and knowledge of a plant's growth stages is advantageous as management practices (such as irrigation, pest and disease control and pruning) can be applied at optimum times in the vines growth cycle. Additionally, information regarding growth stages can be useful in forecasting crop yields and even quality. The information gathered from automated, fast, accurate image analysis from in-field images could be used to design, train and validate simulations and forecasting models of vineyards and grape phenology. This type of research is at the forefront of climate research in which novel methods are required to monitor spatio- temporal physiological responses of food sources to changes in the environment. The use of image processing to enhance practices and forecast the success and or quality of wine grapes is in its infancy. While much of the work to date is promising we have not yet achieved the vineyard of the future, where image analysis (image and video) is a powerful tool that is adopted by viticulturists to inform the management of their vineyards.