Table 6 An analytical study on examples of crop maturity monitoring (fruits detection and counting) approaches; highlights the applied algorithm, plant and data type, data pre-processing and the extracted features
References | Application | Algorithm | Plant | Data type | Data pre-processing | Extracted features |
|---|---|---|---|---|---|---|
[123] | Fruit detection | EM | Tomato | High spacial resolution sensor images | Noise and stalks removing spacial segmentation | Shape and size |
[23] | Fruit detection | ANN | Apple fruit and tree canopy | Digital images | Segmentation | Area of fruitsarea of small fruits cross-sectional area of foliage fruit number total cross-section total cross-sectional |
[108] | Fruit detection | SVM | Coffee | Digital images | Segmentation: homogeneous information | 42 colours features |
[5] | Fruit detection | BC Gaussian | Cherry | Digital images | Segmentation: enhancements and specular reflections removing by inward interpolation method | Colours features:RGB |
[52] | Fruit detection &classification | CNN | Strawberries | Digital images | Hand marking regions of interests | / |
[75] | Immature fruit detection | ANN | Peach | Digital images | Hue-Saturation-Intensity for illumination enhancementpixels’ normalisation histogram equalisation reconstruction of images backgoud elimination | Texture features |
[117] | Fruit counting | CNN | Sweet pepperrock melon strawberry apple, avocado mango, orange | Multi-spectral images(RGB,NIR) | Pixel-wise segmentaion bounding box annotation | Colour and texture features |
[122] | Immature fruitcounting | SVM | Green citrus | Digital images | Images conversion from RGB to graycircular Hough transform | 13 texture features |