Diseases affect crops and the resulting productivity; thus, losses in crops mean food losses, which threatens food security. It is therefore important to take measures that enhance early and correct identification of these diseases so that proper measures can be taken to manage them appropriately. The conventional approach in detecting crop diseases is through visual assessment which is most likely to be characterized by several drawbacks such as being time-wasteful and labor intensive. This paper aims to establish the relation as well as comparison between image processing techniques and machine learning model where a more efficient approach to crop disease detection can be implemented. The developed method uses image processing techniques to identify the important features from the images of crops and then uses intelligent algorithms for disease classification. Several empirical studies prove it to be useful and more accurate with a shorter execution time when compared with other approaches. This research seeks to establish how fusion of state-of-the-art image processing and artificial intelligent systems may help detect crop diseases in a way that would help improve agricultural yield and food security.