Abstract:

The image segmentation threshold is acquired to use gray level images information of pixels and histogram in order to segment between backgrou
nd and objects quickly and accurately. An IABCQ(Improved Artificial Bee Colony Algorithm Based on Quantum） in image threshold segmentation is proposed based on the quantum operation and basic ABC(Artificial Bee Colony) algorithm's mechanism. Firstly, this novel algorithm brings qubits probability amplitude's sinusoidal component into the encoding, then adjust the phase angle to update the qubits probability amplitude, which makes the employed bees move to the optimal nectar source to avoid the algorithm's search blindness. Secondly, the chromosomes sine and cosine components are exchanged by quantum non gate so that followed bees nectar source can update complementarily. Thirdly, the limitation in the artificial bee colony is applied so as to avoid the local optimal solutions and fixed point. Finally, many types of images and algorithms comparison can verify that convergent speed of this novel method in image threshold segmentation is reduced about 20% and this algorithm has good stability and anti-noise ability.

Key words: threshold segmentation, quantum, probability amplitude;bee colony algorithm