An Automated System for Counting Achenes on Strawberries

Abstract

Development of strawberry (Fragaria ×ananassa Duch.) fruit depends on the number of fertilized achenes on its surface (Nitsch, 1950). The achenes are distributed on the receptacle surface in a pattern of more or less regular rows, spirally arranged (Takeda et al., 1990). The total number of achenes per fruit is determined during flower bud initiation and depends on fruit hierarchy on the inflorescence, cultivar, and environmental factors (Webb et al., 1978). A quadratic relationship exists between fruit size and fertilized achenes (Khanizadeh et al., 1993). The total number of achenes per fruit is of great importance to strawberry breeders. The number of achenes may be determined by 1) counting the number per square centimeter of surface on ripe fruit (Abbott et al., 1970; Abbott and Webb, 1970); 2) weighing the achenes after separation from the receptacle; or 3) counting the number of achenes after pressing the fruit between two layers of glass (Khanizadeh, 1983). These methods are laborious and time-consuming. Takeda et al. (1990) estimated total achene counts using a regression line based on the number of carpels in a spiral. However, this method only can be used to study the relationship between fruit weight and achene count if all the carpels are fertilized and may not be suitable for malformed fruit that are partially pollinated and may have many aborted achenes. We, therefore, tested a new method as an alternative to the methods noted. An image analysis system (IAS) was developed by Vigneault et al. (1992a) for measuring the mean diameter of air bubbles in a water tank. This system consists of a monochrome camera that produces a video signal. The frame grabber in the PC digitizes the image in a 480 × 512