Single-layer perceptron artificial visual system for orientation detection

Abstract

Orientation detection is an essential function of the visual system. In our previous works, we have proposed a new orientation detection mechanism based on local orientation-selective neurons. We assume that there are neurons solely responsible for orientation detection, with each neuron dedicated to detecting a specific local orientation. The global orientation is inferred from the local orientation information. Based on this mechanism, we propose an artificial visual system (AVS) by utilizing a single-layer of McCulloch-Pitts neurons to realize these local orientation-sensitive neurons and a layer of sum pooling to realize global orientation detection neurons. We demonstrate that such a single-layer perceptron artificial visual system (AVS) is capable of detecting global orientation by identifying the orientation with the largest number of activated orientation-selective neurons as the global orientation. To evaluate the effectiveness of this single-layer perceptron AVS, we perform computer simulations. The results show that the AVS works perfectly for global orientation detection, aligning with the majority of physiological experiments and models. Moreover, we compare the performance of the single-layer perceptron AVS with that of a traditional convolutional neural network (CNN) on orientation detection tasks. We find that the single-layer perceptron AVS outperforms CNN in various aspects, including identification accuracy, noise resistance, computational and learning cost, hardware implementation feasibility, and biological plausibility.

Keywords: computer vision; orientation detection; perceptron; single-layer; visual system.

Figure 1

(A) McCulloch-Pitts neuron model; (B) a single-layer perceptron.

Figure 2

The perceptrons for the four types of orientation-selective neurons in a 3 × 2 local receptive field. (A) 0°-selective neuron, (B) 45°-selective neuron, (C) 90°-selective neuron, and (D) 135°-selective neuron.

Figure 3

The neural connections in a local receptive field, (A) the connections between photoreceptors and orientation-selective neurons, (B) the perceptron form of the connections between photoreceptors and orientation-selective neurons.

Figure 4

The mechanism of the single layer perceptron AVS for two-dimensional global orientation detection.

Figure 5

The single-layer perceptron AVS for two-dimensional global orientation detection.

Figure 6

Simulated responses of the local orientation detective neurons to a line stimulus of 1 × 10 at a 135° orientation (A), overall activations (B) and individual activations of 0°-selective neurons, 45°-selective neurons, 90°-selective neurons, and 135°-selective neurons (C).

Figure 7

Simulated responses of the local orientation detective neurons to a bar stimulus of 4 × 10 at a 135° orientation (A), overall activations (B), and individual activations of 0°-selective neurons, 45°-selective neurons, 90°-selective neurons, and 135°-selective neurons (C).

Figure 8

The architecture of the single-layer perceptron AVS (A) and CNN (B) used in experiments.

Figure 9

Learning results of loss (A) and accuracy (B) of the CNN.