Abstarct: The helmet is an essential equipment in motorcyclists, which can reduce the damage for the head area during an accident. However, sometimes, some motorcyclists do not use it for various reasons. For this purpose, the law of using helmet in motorcyclists is laid down in many countries. The use of human inspectors to detect a violation of the law is a continuous and boring along with error task. Therefore, intelligent systems are needed to detect the violations. In the past few years, to detect motorcycle riders without helmet, researches were developed using image processing strategies. In the methods, to identify the helmet, low-level (color, shape, and texture) or high level (semantic features) features are used. Most researches in the extraction of low-level features in the motorcyclist's head used several descxriptors. Using of several descxriptors takes a lot of calculations and computational time. Also, a system that uses high-level features requires a lot of data for training. In the proposed method, an approach is provided to diagnose motorcyclists without helmets, which extracts low-level features from motorcyclists' images (head area) using image processing techniques. The proposed method uses the concept of texture to extract the feature for recognizing the helmet in image processing. By analyzing the head image and its texture, it can properly detect motorcyclists without helmet. To improve the accuracy of the method, in addition to the concept of texture, the geometric shape feature (gradient of the local area of the image) is also used to identify the helmet. To evaluate performance of the proposed method, it was applied on a databaxse containing 255 images of motorcyclists with and without helmet. The best results are obtained using the texture, and texture along with the geometric shape method are 95.86% and 98.03%, respectively. The method was also compared with an existing method on a databaxse in terms of accuracy and speed; and the results indicate the superiority of the proposed method.