257 results about "Image classifier" patented technology

Methods and arrangements involving portable devices, such as smartphones and tablet computers, are disclosed. In a particular embodiment, a system stores information from a sensor sub-system as RDF triples. The sensor sub-system may comprise a physical or logical sensor, such as a camera, a microphone, an accelerometer, a GPS receiver, an image classifier, and a user activity sensor. The triples can be stored in a user's smartphone or other portable device, or in the cloud. The stored data can then be acted on by a processor—again, in the user's smartphone, in another portable device, or in the cloud. A great variety of other features and arrangements are also detailed.

The invention discloses a system and a method for filtering sensitive webpage, which is based on multi-classifier fusion. The processing object is a webpage, and the processing result is whether the webpage contains sensitive content, which may be pornography, reaction, violence and other unhealthy Internet contents harmful to society. The system comprises a data stream obtaining and preprocessing unit, an image and text stream filtering unit and an information fusion unit of image filter and text filter, by the cooperation of multiple classifiers, the system acquires source code of a webpage by using the URL of the webpage, a text and an image are separated at preprocessing stage to obtain text information and effective image information; an input webpage is divided into three modes by decision tree algorithm; the webpage is recognized by using a consecutive text classifier, a discrete sensitive text classifier and an image classifier, the output result recognized by the classifiers is fused and calculated, then a judge factor is given, and the final result is returned to a browser.

A method for segmenting and measuring anatomical structures in fetal ultrasound images includes the steps of providing a digitized ultrasound image of a fetus comprising a plurality of intensities corresponding to a domain of points on a 3-dimensional grid, providing a plurality of classifiers trained to detect anatomical structures in said image of said fetus, and segmenting and measuring an anatomical structure using said image classifiers by applying said elliptical contour classifiers to said fetal ultrasound image, wherein a plurality of 2-dimensional contours characterizing said anatomical structure are detected. The anatomical structure measurement can be combined with measurement of another anatomical structure to estimate gestational age of the fetus.

The present invention relates to a method for defense of attack of adversarial examples based on a convolutional denoising auto-encoder. Adversarial image examples x* (output tags identified by an image classification device is y*) are constructed by manual addition of adversarial disturbance on clean image samples x without modification (output tags identified by the image classification device is y), the fraud purpose that y* is not equal to y can be achieved, even though the image classification device classifies two images essentially showing the same meaning to two classes by mistake. Thepresent invention designs an integration defense model connected with a target image classifier based on a convolutional denoising auto-encoder (CDAE), namely input samples are subjected to coding and decoding at the internal portion of a well trained CDAE to remove most of adversarial disturbances in the input samples so as to output denoising samples close to original clean samples, and then are transmitted to the target image classifier so as to improve the classification correction of the target classifier and have an effect for defense of attack of adversarial examples.

The invention discloses an image classification method based on a visual dictionary and relates to the technical field of digital image processing. The image classification method comprises the following steps of: 1, extracting a union partial characteristic of a training image data set; 2, performing vector vectorization on the union partial characteristic by using a clustering algorithm based on a moving mean value and a regional hash method so as to select the number of clustering centers and form the visual dictionary; 3, generating a characteristic expression of images according to the visual dictionary so as to build an image classifier; and 4, classifying the images in the training image data set according to the image classifier. By the image classification method, the visual dictionary having the discrimination can be obtained, so the classification method is adaptive to the sample space distribution of the image data set, high in resistance of affine transformation and lighting variation, robustness to partial abnormity, noise interference and complicated backgrounds, universality and practical value, and can be applied to classification of various images.

The invention discloses an image classifier adversarial attack defense method based on disturbance evolution. The method comprises the following steps that 1) the sample is attacked by different attack models so as to obtain different types of disturbance; 2) a block box model is attached by the adversarial sample corresponding to the disturbance and the attack effect is ordered; 3) cloning, crossing, variation and other operation are performed on the disturbance of great attack effect so as to obtain the new disturbance; 4) the disturbance is updated by using the parent-child hybrid selectionmode so as to achieve the objective of disturbance evolution; 5) an adversarial sample detector is trained by the adversarial sample corresponding to the evolved disturbance and the normal sample; and 6) when the detection sample is detected, detection is performed by using the adversarial sample detector firstly and then the normal sample is inputted to the black box model and the class is returned so as to achieve the adversarial attack defense effect.

The invention discloses a foggy-day driving visual enhancement and visibility early warning system and method based on multiple sensors, and belongs to the technical field of intelligent vehicles safety assistance driving. The system comprises a power supply, a variable-voltage plug, a preposed infrared camera, a preposed millimeter-wave radar, a vehicle-mounted electric control unit module, a vehicle-mounted display screen, a vehicle sound equipment, a vehicle-mounted loudspeaker, and a vehicle speed sensor. Through building a road image classifier and a fog image classifier, building a fog image defogging model and obtaining a clear defogging image, the system and method achieve the effect of visual enhancement, construct a visibility calculation model, judge the visibility level, measure the vehicle speed level through employing the millimeter-wave radar, detect the level of an interval with a front vehicle, and judges whether the visual and acoustic early warning is provided for a driver or not. The system and method can achieve the visual enhancement and driving visibility early warning for the driver under the conditions of a foggy day and low visibility, and solve problems that a conventional vehicle fog light is limited in irradiation range in a foggy data and has a blind region and the driver does not judge the driving environment accurately.

The invention provides an image classification method, an image classification device, an image retrieval method and an image retrieval device, wherein the image classification method concretely comprises the following steps that physical characteristics of images to be classified are extracted; the images to be classified are subjected to semantic annotation to obtain corresponding annotation words; by aiming at the annotation words of the images to be classified, the annotation words are matched with semantic words in a semantic network, and in addition, semantic characterization polybasic groups are generated according to the semantic characterization corresponding to the successfully matched semantic words; semantic words and a plurality of corresponding semantic characterizations are stored in the semantic network, and the semantic characterizations are described by physical characteristics; and characteristic vectors consisting of the physical characteristics of the images to be classified and the semantic characterization polybasic groups are input into an image classifier, and corresponding classification results are output, wherein the image classifier is a classifier obtained according to the image sample training under each image type, and the physical characteristics and the arity of the semantic characterization polybasic groups are identical in the training and classification process. The methods and the devices provided by the invention have the advantage that the image classification accuracy can be improved.

The invention provides an image text description method based on a knowledge transfer multi-modal recurrent neural network. According to the method, by a knowledge transfer model in a multi-modal unit, identification capacity of an existing image classifier on most of objects and grammatical structures and semantic association in an existing corpus are well utilized, a target object in an image can be more accurately described, and generated sentence descriptions can be richer in grammatical structure, appropriate in semantics, and higher in readability.

The invention discloses an image classification method and system based on an image salient region. The method includes offline training and online test. The offline training comprises: performing ultra-pixel segmentation on an image to obtain multidimensional segmentation blocks, and calculating the characteristic contrast of the segmentation blocks to obtain a target salient map; performing threshold segmentation on the target salient map to obtain a binary image, performing morphological processing on the binary image, and performing automatic segmentation extraction on the target salient map by employing a segmentation algorithm to obtain the salient region; and inputting the salient region to a convolutional neural network for training to obtain an image classifier based on the image salient region. The online test includes: performing automatic segmentation extraction of the salient region on a test image, inputting a salient region image of the test image to the trained image classifier, and performing image classification to obtain an image class mark. According to the method and system, the segmentation result is guaranteed, the workload of artificial interaction is reduced, and the accuracy of image classification is improved.

The invention discloses an undesirable image detecting method based on connotative theme analysis, which is substantially used for solving the problem of wrong judgment on normal images resulting from semantic information consideration failure in the present undesirable information detecting method. The scheme is as follows: extracting a skin region of an image by a double-blending Gaussian model; generating a codebook base containing distinguishing features in the skin region by a word bag model, and representing each training image to a group of word co-occurrence vectors with weights via aword frequency-inverse identification file frequency method; forming all co-occurrence vectors to a co-occurrence matrix, performing LDA model creation on the co-occurrence matrix to obtain the themeof the image; inputting the mixed theme of the training image in a BP neural network to train an undesirable image classifier; and obtaining the theme of an image to be measured, inputting the theme to the undesirable image classifier, and judging whether the theme is an undesirable image so as to finish the undesirable image detection. As shown in the test, the invention can be used for better distinguish the undesirable images and the normal images, so that the invention can be used for filtering the erotic information in the images.

The invention discloses an image content identification method and apparatus. The method comprises: extracting a CNN feature of a to-be-identified image; inputting the to-be-identified image into a pre-established image classifier set, obtaining a probability value output by each image classifier in the image classifier set according to the CNN feature of the to-be-identified image, and selecting at least one image classifier as a target image classifier from the image classifier set according to the probability value output by each image classifier; determining the similarity between each image in the target image classifier and the to-be-identified image, and selecting multiple images as target images from the target image classifier according to the similarity; and marking the to-be-identified image by using a type tag corresponding to a class cluster in the target image classifier and an entity tag corresponding to the target image in the target image classifier. With the adoption of embodiments of the invention, the accuracy of image identification can be improved.

The invention provides an Internet picture filtering method and device. The Internet picture filtering method comprises the following steps: according to different preset dimensions, zooming an obtained Internet image to generate pictures of different dimensions; calculating the quick characteristics of the pictures of different dimensions; adopting preset image annotation information and the quick characteristics to train to generate a target image classifier; utilizing the target image classifier to detect the Internet image, and determining a candidate image area which contains target attributes; inputting the candidate image area into a convolutional neural network to calculate the characteristics of the convolutional neural network; and utilizing the classifier to classify whether the picture contains a target image or not according to the convolutional neural network to realize picture filtering. Internet pictures can be favorably filtered, calculation efficiency is improved, a user can carry out fine tuning on a deep neural network on line, and therefore, detection performance of the Internet picture filtering method is better than the detection performance of other methods.

An image similar to a target image is selected from among a set of candidate images. A set of image classifiers is first generated and used to create a fingerprint for each candidate image. A hash table is generated for each fingerprint segment, and an identifier for each candidate image is stored in each hash table based on the candidate image fingerprint value for the fingerprint segment associated with the hash table. A fingerprint is created for the target image using the set of classifiers. Segments of the target image fingerprints are compared to segments of the candidate image fingerprints using the hash table, and a candidate image similar to the target image is selected based on this comparison.

The invention relates to a pornographic image recognizing method comprising: a first step of establishing a sample image database; a second step of extracting characteristics of sample images; a third step of training the characteristics extracted from the sample images to form a pornographic image classifier; a fourth step of extracting the characteristics of an image needing to be recognized; and a fifth step of inputting the characteristics of the image needing to be recognized in the pornographic image classifier, thus obtaining a recognition result. The image characteristics employ a color histogram characteristic vector consisting of determined values, fuzzy values and intuition values together. The invention further comprises a step of optimizing the image characteristics by genetic algorithm. When in use, the method can increase the image recognition accuracy and shorten the recognition time.

The present invention discloses a satellite remote sensing image cloud amount calculation method on the basis of random forest. The satellite remote sensing image cloud amount calculation method on the basis of random forest comprises six steps: sample acquisition, feature extraction, image classifier training, segmentation of image to be measured, image classification, cloud amount calculation and the like. Through adoption of the method provided by the invention, multiple detections may be performed after training just once, an image classifier is obtained through a large number of image trainings, and the image classifier may be used again when cloud detection is performed. The random forest algorithm is low in time complexity at the prediction classification stage, and the cloud zone detection may be rapidly carried out. Through the test, the method provided by the invention is applicable to panchromatic images (ten-dimensional characteristic vector) and also applicable to n-channel multispectral images (10n-dimensional characteristic vector), and has been applied to an actual quality control system of satellite image products, so that the cloud detection of remote sensing images of multiple domestic satellites such as the resource satellite-3, mapping satellite-1, GF-1 and the like are performed, wherein the accuracies reach, respectively, 91%, 88% and 92.4%.

The invention discloses an incremental learning image classification training method under the big data scene. The method comprises steps that step 1, an initial image classifier is trained through utilizing original image data; step 2, if a new class image appears in an application process, incremental training of an initial model is carried out, and the image classifier after update is acquired; and step 3, to-be-classified images are identified through utilizing the trained image classifier, and test image classes are acquired. The method is advantaged in that the convolutional neural network is utilized to carry out image characteristic extraction, manual characteristic definition can be avoided, and the high identification degree is realized; the convolutional neural network is utilized to carry out image identification, memory occupation is small, the calculation speed is fast, incremental leading of the new class image can be carried out, storage of the original training data is not needed, the training time and the storage time can be substantially saved, and the method is especially suitable for big data image classification occasions.

The invention discloses an instantaneous myoelectricity image based gesture identification method. During a training stage, firstly instantaneous myoelectricity signals acquired by array electrodes are preprocessed and arranged according to electrode positions to form an instantaneous myoelectricity image; and secondly an image classifier such as a deep convolutional neural network is trained by using the instantaneous myoelectricity image and a gesture tag corresponding to the instantaneous myoelectricity image to obtain network model parameters. During a test stage, firstly to-be-identified instantaneous myoelectricity signals acquired by the array electrodes are preprocessed and arranged according to the electrode positions to form the instantaneous myoelectricity image; and secondly the trained model parameters are substituted into the classifier to identify the gesture tag corresponding to the instantaneous myoelectricity signals. According to the instantaneous myoelectricity image based gesture identification method, a gesture can be quickly and accurately identified based on the instantaneous myoelectricity image and an image classification method. No literature for gesture identification by the instantaneous myoelectricity signals exists at home and abroad yet.

A method, apparatus and computer program product are hereby provided to detect anatomical elements in a medical image. In this regard, the method, apparatus, and computer program product may receive a test image and generate a classified image by applying an image classifier to the test image. The image classifier may include at least one decision tree for evaluating at least one pixel value of the test image and the classified image may include a plurality of pixel values. Each pixel value may be associated with a probability that an anatomical element is located at the pixel location. The method, apparatus, and computer program product may also evaluate the classified image using an anatomical model to detect at least one anatomical element within the classified image.

A blood cell subtype image classification method based on multi-scale fusion is provided. The method comprises the following steps: (1) wherein a training set contains four erythrocyte subtype images;(2) based on an Xception Entry flow module, constructing a shallow feature extraction network, and outputting four feature maps of different scales; (3) respectively connecting four intermediate feature extraction networks formed by cascading Xception Middle flow modules for the four outputs in step (2); (4) respectively connecting four modified Xception Exit flow modules for the four outputs instep (3), extracting deep feature information, and outputting four high-dimensional feature vectors; (5) fusing the information of 4 high-dimensional feature vectors output in (4), and performing inference and prediction on image categories; (6) on the classification network, using the data-amplified blood cell subtype image training set for training; and (7) using the network trained in step (6)to predict on a blood cell subtype image test set, and outputting categories to which the blood cell subtype images belong. According to the method disclosed by the present invention, the performanceof a blood cell subtype image classifier can be enhanced.

The invention discloses a support vector machine-based cloud, snow and fog detection method for an optical satellite remote sensing image. The method comprises the following steps of firstly, collecting different types of a large amount of ground object and cloud, snow and fog sample image data to serve as a training set, obtaining grayscale and texture features of images to form a feature set, and performing machine learning on the feature set of all samples through a support vector machine method to obtain cloud, snow and fog image classifiers; secondly, determining the types of the to-be-detected images by using the obtained cloud, snow and fog image classifiers, performing morphological close operation and overlapping region correction, and judging the type of a target region in the remote sensing image; and finally, re-selecting training samples to obtain new image classifiers, performing secondary detection on the to-be-detected remote sensing image, and performing comparison with first detection to finally determine cloud, snow and fog judgment results of the to-be-detected remote sensing image. An experimental result shows that the method can achieve relatively high detection precision.

The invention discloses a video content detection method and a video content detection system based on self-adaption sampling. The video content detection method includes: establishing a training graph set, training according to the training graph set to obtain an image classifier, and subjecting a to-be-detected video to decoding to generate an image sequence; based on the self-adaption sampling method, utilizing the image classifier to perform sampling detection upon the generated image sequence, and according to a result of the sampling detection of the image sequence, judging whether the to-be-detected video is an objectionable video or not. Through the technical scheme and adoption of the dynamic self-adaption sampling method, when video content is detected, good balance is achieved in detection accuracy and detection efficiency, the detection accuracy is guaranteed, number of sampling frames is decreased, the detection efficiency is improved, and beneficial effect of rapid detection can be achieved.

A system and method for detecting differences in images and identifying groups of images with similar differences detected is provided. To this end, an image comparator may be operably coupled to a pixel comparator for comparing pixels of the baseline image with corresponding pixels of one or more sample images to detect differences between the pixels compared. A derivative difference processor may be also provided which may be operably coupled to the pixel comparator for determining a difference region of pixels detected as different by the pixel comparator in the sample images. An image sorter may also be provided which may be operably coupled to the difference derivative processor for grouping together sample images that have a similar difference region determined by the derivative difference processor.

The invention discloses an image classifier establishing method and an image classifier establishing device. The image classifier establishing method comprises steps that a sample image set is acquired, and the sample image set comprises a positive sample comprising a target image and a negative sample without the target image; the deformation processing of the sample images in the sample image set is carried out to acquire the expanded sample image set; according to the expanded sample image set and a deep convolutional neural network model, a classifier aiming at the target image is acquired; the normalization processing of the output of the convolutional layer of the deep convolutional neural network is carried out. By adopting the above mentioned method, the limited sample image sets are identified artificially, and then the limited sample image sets are expanded, and therefore a sample amount is expanded, and the accuracy of the classifier is improved. The identification accuracy of the classifier is further improved by adopting the deep convolutional neural network, and therefore the classifier aiming at the target image and having the higher identification accuracy is provided.

A processing device for performing visual recognition using deep learning attributes and method for performing the same are described. In one embodiment, a processing device comprises: an interface to receive an input image; and a recognition unit coupled to the interface and operable to perform visual object recognition on the input image, where the recognition unit has an extractor to extract region proposals from the input image, a convolutional neural network (CNN) to compute features for each extracted region proposal, the CNN being operable to create a soft-max layer output, a cross region pooling unit operable to perform pooling of the soft-max layer output to create a set of attributes of the input image, and an image classifier operable to perform image classification based on the attributes of the input image.

A method and system for image processing, in conjunction with classification of images between natural pictures and synthetic graphics, using SGLD texture (e.g., variance, bias, skewness, and fitness), color discreteness (e.g., R_L, R_U, and R_V normalized histograms), or edge features (e.g., pixels per detected edge, horizontal edges, and vertical edges) is provided. In another embodiment, a picture / graphics classifier using combinations of SGLD texture, color discreteness, and edge features is provided. In still another embodiment, a “soft” image classifier using combinations of two (2) or more SGLD texture, color discreteness, and edge features is provided. The “soft” classifier uses image features to classify areas of an input image in picture, graphics, or fuzzy classes.

An image processing apparatus, includes an image classifier configured to determine whether an input image is a low-quality image or a high-quality image; and an image evaluator configured to determine a first predetermined number of clearest images from a plurality of low-quality images determined by the image classifier.