Abstract: A camera device with a uniform illumination effect includes a base, a plurality of camera units, a light source and a covering component. The plurality of camera units is disposed on the base. The light source is located among the plurality of camera units. The covering component is disposed on the base to cover the plurality of camera units and the light source. The covering component includes a central area and an outer ring area. The outer ring area is disposed around the central area and corresponds to the plurality of camera units. The central area is integrally formed with the outer ring area and corresponds to the light source.

Abstract: An image content analysis method is applied to an image analysis apparatus and includes acquiring an image, utilizing an edge detection technology to compute an edge density of the image, utilizing a texture detection technology to compute a texture richness of the image, and analyzing the edge density and the texture richness to generate a richness score of the image.

Abstract: An image analysis method is applied to a surveillance apparatus and includes an operation processor of the surveillance apparatus setting a first object bonding box within a surveillance image, the operation processor acquiring data of a candidate bonding box within the surveillance image via a first object threshold, the operation processor computing an intersection parameter of the first object bonding box and the candidate bonding box, and the operation processor deciding a first object decided threshold in accordance with the intersection parameter.

Abstract: A natural language image searching method is applied to an image searching system including an image analysis unit, a data management unit and an instruction input unit. The natural language image searching method includes an image feature vector encoder of the image analysis unit receiving a detection image and generating an image feature vector, the image analysis unit transmitting the detection image, the image feature vector, and a time stamp associated with the detection image to the data management unit, a text encoder of the instruction input unit generating and transmitting a text feature vector to the data management unit in accordance with a query statement, and the data management unit determining whether the query statement generated by the instruction input unit via natural language conforms to the detection image in accordance with a comparison result between the text feature vector and the image feature vector.

Abstract: A resolution detection device is applied to a camera device and includes at least one arc-shaped structural member and an identification member. The at least one arc-shaped structural member is disposed on a position distant from a detection point in a predefined distance. The identification member includes a plurality of identification patterns. The identification member is formed on the at least one arc-shaped structural member, and a plurality of intervals between the detection point and each of the plurality of identification patterns is set as the predefined distance. The camera device acquires a captured image via the detection point, and the captured image contains the plurality of identification patterns. The resolution detection device further includes a relay lens disposed on the detection point.

Abstract: An image analysis method is applied to a camera apparatus and used to analyze the landform inside a detection image captured by the camera apparatus via a predefined installation angle. The image analysis method includes dividing the detection image into a plurality of first regions to acquire a plurality of first height, defining some first regions with the same height and adjacent to each other as a first reference plane area to acquire a first plane formula, selecting at least one first region adjacent to and outside the first reference plane area, applying the selected first region for the first plane formula to acquire a first planar height and comparing the first planar height with an initial first height of the at least one first region, and determining landform between the first reference plane area and the selected at least one first region in accordance with a comparison result.

Abstract: An image quality improvement method is applied to an operation processor of a surveillance system and includes driving a first light source of the surveillance system to emit a first illumination beam, driving an image camera of the surveillance system to capture a first image relevant to the first illumination beam, and deciding whether to drive a second light source of the surveillance system to emit a second illumination beam and/or decide an illumination intensity of the second illumination beam in accordance with an analysis result of the first image.

Abstract: An adjustable seat includes a mount, a rotary base, a rotary fixing frame, a fixing frame transmission mechanism, and a fixing frame driving module. The rotary base is rotatably connected to the mount and is rotatable around a first rotation axis relative to the mount. The rotary fixing frame is rotatably connected to the rotary base and is rotatable around a second rotation axis relative to the rotary base. The fixing frame transmission mechanism is rotatably connected to the rotary fixing frame. The fixing frame driving module is disposed on the mount and connects with the fixing frame transmission mechanism. The fixing frame driving module drives the rotary fixing frame through the fixing frame transmission mechanism to rotate around the second rotation axis relative to the rotary base. A camera includes the adjustable seat and an image-capturing module fixed on the rotary fixing frame for capturing images.

Abstract: An image recognition method is applied to an image recognition device and used to determine whether the same target object is existed in different images. The image recognition method includes analyzing two continuous images of an image stream to respectively search two target objects and acquire a plurality of feature vectors of the two target objects, computing two distances of the two target objects respectively relative to one reference point of the two continuous images, utilizing the two distances to acquire a corresponding weight of partial feature vectors of the plurality of feature vectors, and utilizing the corresponding weight to adjust the partial feature vectors for generating similarity of the two target objects.

Abstract: An image calibration method is used to calibrate position offset between a first image and a second image captured by a surveillance apparatus in different zoom modes. The image calibration method includes acquiring a plurality of first feature points of the first image and a plurality of second feature points of the second image, dividing the plurality of first feature points and the plurality of second feature points at least into a first group and a second group according to distribution density of the plurality of first feature points and the plurality of second feature points, deciding whether the first group corresponds to the second group, and utilizing coordinates of the first group and coordinates of the second group to compute a position shifting value when the first group corresponds to the second group.

Abstract: An object sampling method is applied to an image analysis apparatus with an image receiver and an operation processor. The image receiver acquires an image stream. The object sampling method includes transforming color space of at least one surveillance image in the image stream, computing saturation and gradient of specific pixels within the at least one surveillance image, setting a bonding box within the at least one surveillance image, and computing a background score relevant to the saturation and value of all pixels in the bonding box, and comparing the background score with a first sampling condition so as to determine whether the at least one surveillance image is applicable for the image analysis apparatus via a comparison result of comparing the foreground score with the first sampling condition.

Abstract: A calibration method of apparatus installation parameter is applied to a surveillance device having an image receiver and an operation processor. The calibration method includes analyzing a detection image from the image receiver to acquire an apparatus installation parameter of the surveillance device, computing a height of at least one target object inside the detection image in accordance with the apparatus installation parameter, analyzing the height of the at least one target object via an error function to acquire an error computed value, computing a local minimal value of the error computed value relevant to the apparatus installation parameter via a gradient descent function, and finding the local minimal value to acquire an installation parameter vector so that the surveillance device executes analysis of the detection image according to the installation parameter vector.

Abstract: An object tracking method is applied to a surveillance apparatus and includes acquiring a first image, a second image and a third image in sequence, at least detecting a first object in the first image and at least detecting a second object in the second image, deciding the first object and the second object are relative and then classify a static object or a dynamic object, at least detecting a current object in the third image, utilizing a low loading tracking algorithm to compare the static object with the current object, and deciding whether the current object is suitable for the low loading tracking algorithm according to a comparison result.

Abstract: An adaptive frame dropping control method is applied to an image processing system. The adaptive frame dropping control method includes utilizing a drop control module of the image processing system to acquire a first image without image processing from a latency buffer, utilizing the first image to compute a first average consuming period of the latency buffer, acquiring a second image without the image processing and later than the first image from the latency buffer for computing a left period of the second image, and determining whether to drop the second image in accordance with a comparison result of the left period and the first average consuming period.

Abstract: An image analysis method of increasing identification efficiency is applied to an image analysis device having an image receiver and an operation processor. The image analysis method includes setting a target pixel per feet (PPF) and detecting a specific area within a surveillance image acquired by the image receiver, computing a first dimension ratio difference between the target PPF and an initial PPF of the specific area, utilizing the first dimension ratio difference to adjust the specific area so that a second dimension ratio difference between the target PPF and an adjusted PPF of the adjusted specific area conforms to a preset condition, and utilizing the adjusted specific area with the adjusted PPF to be detection data for object detection network.

Abstract: An image restoration method is applied to an image restoration device and includes acquiring an analyzing image containing a region of interest and a background region, utilizing a first restoration function to calibrate the region of interest and generate a calibrated region of interest, and combining the calibrated region of interest with the background region via a relative position between the region of interest and the background region for providing a restoration image.

Abstract: An image analysis model adjustment method is applied to an image analysis apparatus. The image analysis model adjustment method includes dividing a detection image acquired by the imager analysis apparatus into a plurality of first sub-images via a valid size, transforming the plurality of first sub-images from a spatial domain to a frequency domain to generate a plurality of first frequency domain images, distributing the plurality of first frequency domain images into several crop groups via a predefined set value, analyzing frequency responses at the same frequency in each crop group to provide a specific frequency response, collecting specific frequency responses of the crop groups to generate frequency domain group data, and analyzing the frequency domain group data to acquire a prediction result of the detection image so as to decide whether to adjust the valid size in accordance with the predicted result.

Abstract: An object classifying and tracking method is applied to an image stream acquired by a surveillance camera for object identification. The object classifying and tracking method includes acquiring a first classify feature and a first dependent feature of a first target object in a first image of the image stream and a second classify feature and a second dependent feature of a second target object in a second image of the image stream, acquiring a dependent degree of the first classify feature and the second classify feature from a memory, transforming the second dependent feature via the dependent degree, and analyzing the first dependent feature and the transformed second dependent feature to determine whether the first target object and the second target object are the same object.