The invention relates to an Internet porn image detection method based on a deep convolution nerve network. The method comprises the following steps of acquiring a porn image and a normal image through a manual calibration method and carrying out pretreatment and enhancement on the images so as to acquire an effective square training image; sending the acquired effective image into a deep convolution nerve network so as to train the network; verifying a network model on a verification set, adjusting a training set according to a result and continuously training the deep convolution nerve network; repeating the last step till that detection accuracy on the verification set reach an expected object or a network loss function begins convergent; testing the trained network on the training set. The method in the invention has the following advantages that the porn image detection method based on the deep convolution nerve network is provided and the method can be used to rapidly detect almost all the types of porn images through a single model; in an actual test, detection accuracy in the invention reaches above 98.6%.

Internet porn image detection method based on deep convolution nerve network

The invention discloses a novel infrared image focusing system The novel infrared image focusing system is characterized by comprising a focusing lens, a direct-current motor, an image processing device, an infrared detector and a data storage device, wherein the direct-current motor is connected with the focusing lens; the image processing device is connected with the direct-current motor; and the infrared detector and the data storage device are respectively connected with the image processing device Simultaneously, the invention further discloses a use method of the novel infrared image focusing system The use method is characterized by comprising the steps of initializing the focusing lens, driving the focusing lens to move in a large step, simultaneously obtaining an infrared image and the like The novel infrared image focusing system disclosed by the invention is rapid in focusing efficiency and high in credibility; simultaneously, the novel infrared image focusing system is not influenced by environmental illumination by adopting an infrared imaging principle; and thus, the focusing precision is increased

Novel infrared image focusing system and use method thereof

The invention provides a fuzzy kernel obtaining and image de-blurring method and apparatus wherein the method comprises the following steps: mapping a fuzzy picture into a cepstral domain; obtaining the cepstral image of the fuzzy picture; according to the cepstral image, determining the fuzzy category of the fuzzy picture; according to the cepstral image, using the fuzzy kernel size statistical method corresponding to the fuzzy category to calculate the fuzzy kernel size; and based on the fuzzy kernel size, generating the fuzzy kernel. The embodiments of the invention do not need manual intervention to obtain the fuzzy kernel size so that calculations become simpler and more accurate.

Fuzzy kernel obtaining and image de-blurring method and apparatus

The invention provides a spectrum and cepstrum characteristic-based fuzzy image detection method. The method comprises the steps of firstly, dividing an image into small regions, and calculating an energy density distribution value and a singular cepstrum distribution value in the small regions; secondly, assuming that two fuzzy characteristics conform to normal distribution and have relatively large difference for a clear image and a fuzzy image, and estimating distribution parameter values of the two characteristics under fuzzy and clear hypotheses; and finally, building a Naive Bayesian classifier, which is used for judging whether a test image is fuzzy or not when the test image is given.

Naive Bayesian method-based fuzzy image detection method

The invention provides a method for detecting a blurred image based on spectrum and cepstrum features, and uses a deep classification network to calculate the probability of the presence of blur in an image. The method achieves excellent performance. The method proposes to take a vectorized spectrum feature and a vectoized cepstrum feature as the blurred features of an image which are to be used in blurred detection. The method analyzes the blurred features of the image, establishes a deep classification network model, taking the output value of the model as the probability of the image being determined as a blurred image, and when the probability is higher, determines the image to be a blurred image. The method introduces deep learning network, can achieve excellent effects of classification, can detect blurred images without references, and can be widely applied to the field of image identifying.

Method for detecting blurred image based on deep learning

The invention discloses a blurred image detection method fusing frequency spectrum information and cepstrum information, belongs to the technical field of image processing, and particularly relates to the detection technology of various blurred images. According to the blurred image detection method, first, an energy frequency spectrum distribution feature and a singularity cepstrum value histogram feature of an image are calculated, and serve as blur features of the image; second, a support vector machine classifier is selected for differentiating sharp image features from the blur image features, and collected images with demarcated blur categories is used for training corresponding parameters of the support vector machine classifier; finally, the trained support vector machine classifier is used for detecting whether an image to be detected is a blurred image. The blurred image detection method has the advantage that as a non-reference blurred image detection method, the blurred image detection method needs no reference image, thereby being wide in application range; meanwhile, the defined blur features have specific physical significance, and therefore the sharp image and the blurred image can be differentiated accurately.

Zheng Yali

Papers

Blurred image detection method fusing frequency spectrum information and cepstrum information