The advancement of quantum computing offers potential advantages in parallelism and kernel-based methods. Our project applies a Quantum Support Vector Machine (QSVM) for the task of classifying the ripeness of blackberries using computer vision techniques. The traditional fruit ripeness detection methods in agriculture often relay on manual inspection otherwise classical machine learning models. This model requires large datasets and extensive preprocessing. This project proposes a hybrid classical-quantum approach that integrates classical feature extraction from images with quantum-enhanced classification. Images of blackberries are categorized into two classes such as unripe and ripe. Then they are processed using OpenCV to extract HSV-based colour histograms. These histograms capture the color profile that changes significantly as blackberries ripen. To keep the feature space small and suitable for current quantum processors, only the first two normalized histogram values were used. In our QSVM model we use Qiskit’sQuantumKernel with a ZZFeatureMap to transform classical image features into quantum states. This allows the model to measure similarities between samples in a quantum-enhanced way. We trained and tested the model using Qiskit’sqasm_simulator with each quantum circuit of 1024 times to ensure stable results. We evaluated our model with classification reports and confusion matrix.