Software Implementation For Fruits And Vegetables Quality Determination

Human always use their senses to detect the quality of the vegetables and fruits
so then they will know how fresh the vegetables or fruits they want However, human
senses can only detect the freshness at a certain level making it hard for us to know how
fresh the vegetables and fruits are. Even though the physical appearances of the
vegetables and the fruits can easily indicate how fresh they are but it can be very
deceiving sometimes as the fruits can be rotten on the inside. The objectives of this
project are to analyze the waveforms obtained from gas emitted by fruits and vegetables
in certain condition and use neural network to classifY the readings from the sensors
into three different conditions which are fresh, slightly fresh and rotten. C program is
also constructed as another way to classifY the fruits and vegetables into three different
conditions as well. The experiment is conducted by obtaining the readings from the
sensors and analyzed the readings using neural network. Based on the simuJation from
the neural network, the network will identifY the conditions of the fruits and vegetables
based on the readings from the sensors. C program also works similar ways like the
neural network. By comparing both neural network and C program, neural network is
able to determine the freshness of the fruits and vegetables with the accuracy of 99%.
Unlike neural network, C program can only determine the condition of the fruits and
vegetables based on the sensor range which it is not sensitive to the changes in the
readings. Therefore, the best method for determining the freshness of the fruits and
vegetables is by using neural network.