Abstract: Techniques for providing increased support for deduplication and compression of encrypted storage volumes. The techniques include receiving, at a storage virtual machine (VM), a data encryption key (DEK) associated with encrypted volume data, in which the DEK is wrapped in a key encryption key (KEK). The techniques include receiving, at the storage VM from a client virtual machine (VM), a write request specifying the encrypted volume data. The techniques include obtaining, by the storage VM, the KEK from a key management system (KMS) embedded on the storage VM. The techniques include unwrapping, by the storage VM, the DEK using the KEK, and decrypting, by an IO decryptor hosted by the storage VM, the encrypted volume data using the DEK. The techniques include performing, by the storage VM, data reduction operations on the decrypted volume data, and storing, by the storage VM, the data-reduced volume data on a storage array.

Abstract: A target detection method based on fusion of vision, lidar and millimeter wave radar comprises: obtaining original data detected by a camera, a millimeter wave radar, and a lidar, and synchronizing the millimeter wave radar, the lidar, and the camera in time and space; performing a calculation on the original data detected by the millimeter wave radar according to a radar protocol; generating a region of interest by using a position, a speed, and a radar reflection area obtained from the calculation; extracting feature maps of a point cloud bird's-eye view and the original data detected by the camera; projecting the region of interest onto the feature maps of the point cloud bird's-eye view and the original data detected by the camera; fusing the feature maps of the point cloud bird's-eye view and the original data detected by the camera, and processing a fused image through a fully connected layer.

Abstract: Provided is a traffic sign recognition method based on a lightweight neural network, which including: a lightweight neural network model is constructed for training and pruning to obtain a lightweight neural network model; the lightweight neural network model comprises a convolution feature extraction part and a classifier part; the convolution feature extraction part includes one layer of conventional 3×3 convolution and 16 layers of separable asymmetric convolution. The classifier part includes three layers of separable full connection modules.

Abstract: Provided is a traffic sign recognition method based on a lightweight neural network, which including: a lightweight neural network model is constructed for training and pruning to obtain a lightweight neural network model; the lightweight neural network model comprises a convolution feature extraction part and a classifier part; the convolution feature extraction part includes one layer of conventional 3×3 convolution and 16 layers of separable asymmetric convolution. The classifier part includes three layers of separable full connection modules.

Abstract: Disclosed are a defense method and a model of deep learning model aiming at adversarial attacks in the technical field of image recognition, which makes full use of the internal relationship between the adversarial samples and the initial samples, and transforms the adversarial samples into common samples by constructing a filter layer in front of the input layer of the deep learning model; the parameters of the filter layer are trained by using the adversarial attack samples, so as to improve the ability of the model to resist adversarial attack; then the trained filter layer is combined with the learning model after the adversarial training, and a deep learning model with strong robustness and high classification accuracy is obtained, which ensures that the recognition ability of the initial sample is not reduced while resisting the adversarial attacks.

Abstract: Disclosed are a point cloud polar coordinate coding method and a device, including dividing a circular scanning area scanned by a lidar at an equal angle with an angle ?? to obtain a plurality of identical polar coordinate areas; dividing each of the polar coordinate areas with equal length along a radial direction with a length ?r to obtain a plurality of polar coordinate grids and generating a plurality of polar coordinate cylinders corresponding to each of the polar coordinate grids in a three-dimensional space; generating polar coordinate cylinder voxels; extracting structural features from the all point cloud data in each of the polar coordinate cylinder voxels; obtaining a two-dimensional point cloud pseudo-image; boundary supplementing to the two-dimensional point cloud pseudo-image; and performing feature extraction on the two-dimensional point cloud pseudo-image by using convolutional neural networks, and outputting a final feature map.

Abstract: Techniques for adaptive re-keying of encrypted data are provided. For example, a method comprises the following steps. Utilization information associated with a storage system is obtained, wherein the storage system comprises a set of storage devices. The method dynamically selects a re-keying process from a plurality of different re-keying processes based on at least a portion of the obtained utilization information. At least a portion of the set of storage devices are re-keyed in accordance with the selected re-keying process.

Abstract: A target detection method based on fusion of vision, lidar and millimeter wave radar comprises: obtaining original data detected by a camera, a millimeter wave radar, and a lidar, and synchronizing the millimeter wave radar, the lidar, and the camera in time and space; performing a calculation on the original data detected by the millimeter wave radar according to a radar protocol; generating a region of interest by using a position, a speed, and a radar reflection area obtained from the calculation; extracting feature maps of a point cloud bird's-eye view and the original data detected by the camera; projecting the region of interest onto the feature maps of the point cloud bird's-eye view and the original data detected by the camera; fusing the feature maps of the point cloud bird's-eye view and the original data detected by the camera, and processing a fused image through a fully connected layer.

Abstract: An apparatus comprises a processing device configured to receive, at a given data management entity running on a given processing node, a request to create a given cluster of data management entities for a given client. The processing device is also configured to determine membership requirements for the given cluster, to discover additional data management entities running on additional processing nodes, and to select at least one of the additional data management entities for membership in the given cluster based at least in part on the membership requirements. The processing device is further configured to establish a replication relationship for automating sharing of metadata in the given cluster, the metadata comprising access information and location information for data stores where portions of data items of the given client are stored. The processing device is further configured to perform data management functions for the given client utilizing the metadata.

Abstract: A sample protection device for a scanning electron microscope, the sample protection device comprising: a shell; an accommodating part having an accommodating space for accommodating a sample, the accommodating part being arranged in the shell in such a manner that the accommodating part can move relative to the shell, such that the accommodating part at least partially enters the shell or moves out of the shell; a sealing part connected to the accommodating part and configured to seal between the accommodating part and the shell when the accommodating part is at least partially accommodated in the shell; and a driving member configured to drive relative movement of the shell relative to the accommodating part.

Abstract: A technique for backing up snapshots in a multi-cloud environment includes accessing multiple cloud storage providers and obtaining therefrom respective sets of storage performance data. The technique further includes receiving storage requirements for snapshots of a data object and selecting, based on the requirements and the storage performance data, a designated cloud storage provider for receiving archived snapshots of the data object. The technique then transmits one or more snapshots of the data object to the designated cloud storage provider for archiving.

Abstract: A method for managing physical assets includes obtaining a physical asset security assessment request for a set of physical assets, and, in response to the physical asset security assessment request: sending a physical asset initiation request to an administrative client, obtaining a physical asset response from the administrative client, wherein the physical asset response comprises physical asset state information, allocating a security assessment device from a security assessment pool based on the physical asset state information, sending a security assessment process request to the security assessment device, wherein the security assessment process request comprises a security assessment process, the physical asset state information and physical asset information obtained from a physical asset information repository, obtaining a security assessment report from the security assessment device, and after obtaining the security assessment report, sending the security assessment report to the administrative cl

Abstract: A technique rekeys information to maintain data security. The technique involves identifying a first storage drive as a source device available to a proactive copy service. The technique further involves identifying a set of second storage drives as a set of spare devices available to the proactive copy service. The technique further involves invoking the proactive copy service which, in response to being invoked, transfers information from the first storage drive to the set of second storage drives. The information is encrypted by a first key when residing on the first storage drive and is encrypted by a set of second keys when residing on the set of second storage drives, the first key being different from each second key.

Abstract: An apparatus in an illustrative embodiment comprises a client device configured for communication with a storage system, with the client device comprising a processor coupled to a memory. The client device is further configured to identify a data item to be stored in the storage system, and to generate a data encryption key for the data item as a function of a first secret key and the data item. For example, the function may comprise hashing at least the data item. The client device is further configured to encrypt the data item using the data encryption key for the data item, and to send the encrypted data item to the storage system for storage therein. The client device in some embodiments is further configured to encrypt the data encryption key using a second secret key, and to send the encrypted data encryption key to the storage system for storage therein as metadata of the data item.

Abstract: A technique for backing up snapshots in a multi-cloud environment includes accessing multiple cloud storage providers and obtaining therefrom respective sets of storage performance data. The technique further includes receiving storage requirements for snapshots of a data object and selecting, based on the requirements and the storage performance data, a designated cloud storage provider for receiving archived snapshots of the data object. The technique then transmits one or more snapshots of the data object to the designated cloud storage provider for archiving.

Abstract: A technique rekeys information to maintain data security. The technique involves identifying a first storage drive as a source device available to a proactive copy service. The technique further involves identifying a set of second storage drives as a set of spare devices available to the proactive copy service. The technique further involves invoking the proactive copy service which, in response to being invoked, transfers information from the first storage drive to the set of second storage drives. The information is encrypted by a first key when residing on the first storage drive and is encrypted by a set of second keys when residing on the set of second storage drives, the first key being different from each second key.

Abstract: Techniques for adaptive re-keying of encrypted data are provided. For example, a method comprises the following steps. Utilization information associated with a storage system is obtained, wherein the storage system comprises a set of storage devices. The method dynamically selects a re-keying process from a plurality of different re-keying processes based on at least a portion of the obtained utilization information. At least a portion of the set of storage devices are re-keyed in accordance with the selected re-keying process.

Abstract: Techniques for synchronizing configuration information in a clustered storage environment. The techniques allow a system administrator or other user to make additions and/or updates to configuration information in one or more configuration files, which are automatically propagated for storage in multiple data storage appliances within a storage domain. By allowing a user to make changes to configuration files associated with a primary appliance within the storage domain, and automatically propagating the configuration files in a background process from the primary appliance to multiple secondary appliances within the storage domain, the user can more readily assure consistency of the configuration information, not only among the primary and secondary appliances within the storage domain, but also among previously unavailable or unreachable data storage appliance(s) that may be recovered and brought back on line within the storage domain.

Abstract: A challenge/response authentication procedure determines whether a response is a correct response, a unique incorrect response, or a non-unique incorrect response, the unique incorrect response and non-unique incorrect response being differentiated by comparing the response value with a store of unique incorrect response values. For the correct response, client access to protected computer system resources is allowed, and the challenge value is discarded so as not to be used again. For the unique incorrect response, (1) when a predetermined limit of unique incorrect responses has not been reached, then the response value is added to the store of unique incorrect response values and the process is repeated with reuse of the challenge value, and (2) when the predetermined limit has been reached, then the client is locked out. For the non-unique incorrect response, the process is repeated with reuse of the challenge value.

Abstract: An apparatus in an illustrative embodiment comprises a client device configured for communication with a storage system, with the client device comprising a processor coupled to a memory. The client device is further configured to identify a data item to be stored in the storage system, and to generate a data encryption key for the data item as a function of a first secret key and the data item. For example, the function may comprise hashing at least the data item. The client device is further configured to encrypt the data item using the data encryption key for the data item, and to send the encrypted data item to the storage system for storage therein. The client device in some embodiments is further configured to encrypt the data encryption key using a second secret key, and to send the encrypted data encryption key to the storage system for storage therein as metadata of the data item.