Video Enhancement Using Adaptive Spatio-Temporal Connective Filter and Piecewise Mapping

Driven by rapid development of digital devices, camcorders and cameras are no longer used only for professional work, but step into a variety of application areas such as surveillance and home video making. While capturing videos become much easier, video defects, such as blocking, blur, noises, and contrast distortions, are often introduced by many uncontrollable factors: unprofessional video recording behaviors, information loss in video transmissions, undesirable environmental lighting, device defects, and so forth. As a result, there is an increasing demand for the technique—video enhancement, which aims at improving videos' visual qualities, while endeavoring to repress different kinds of artifacts. In this paper, we focus on two most common defects: noises and contrast distortions. While some existing software have already provided noise removal and contrast enhancement functions, it is likely that most of them introduce artifacts and could not produce desirable results for a broad variety of videos. Until now, video enhancement still remains a challenging research problem in filtering noises as well as enhancing contrast.

The natural noises in videos are quite complex; yet, fortunately, most noises can be represented using two models: additive Gaussian noise and impulse noise [1, 2]. Additive Gaussian noise generally assumes zero-mean Gaussian distribution and is usually introduced during video acquisition, while impulse noise assumes uniform or discrete distribution and is often caused by transmission errors. Thus, filters can be designed targeting the two kinds of noises. Gaussian noise can be well suppressed by bilateral filter [3], anisotropic diffusion [4], wavelet-based approaches [5], or fields of experts [6] while maintaining edges. Impulse noise filters lie on robust image statistics to distinguish noise pixels and fine features (i.e., small high-gradient regions) and often need an iterative process to reduce false detection [7–9]. Building filters for removing mixture of Gaussian and impulse noise is more practical than that for one specific type of noise with respect to natural images. The essence of mixed noise filter is to incorporate the pertinent techniques into a uniform framework that can effectively smooth the mixed noise while avoiding blurring the edges and fine features.

As to video noise removal, in addition to the above issues, temporal information should also be taken into consideration because it is more valuable than spatial information in the case of stationary scene [10]. But straightly averaging temporal corresponding pixels to smooth noise may introduce "ghosting" artifacts in the presence of camera and object motion. Such artifacts can be removed by motion compensation and a number of algorithms have been proposed with different computational complexity [11]. However, severe impulse noise will introduce abrupt pixel changes like motions and greatly decrease the accuracy of motion compensation. Moreover, there are often not enough similar pixels for smoothing in temporal directions, owing to imperfect motion compensation or transitions between shots. Thus, a desirable video noise filter should distinguish impulse pixels and motional pixels as well as collect enough similar pixels adaptively from temporal to spatial directions.

As to contrast enhancement after noise filtering, it is quite difficult to find a universal approach for all videos owing to their diverse characteristics such as underexposed, overexposed with many fine features or with large black background. Although numerous contrast enhancement methods have been proposed, most of them are unable to automatically produce satisfactory results for different kinds of low-contrast videos, and may generate ringing artifacts in the vicinity of the edges "washed-out" artifacts [12] when having monochromic background or noise over enhancement artifacts.

Motivated by the above observations, we propose a universal video enhancement system to automatically recover the ideal high-quality signal from noise degraded videos and enlarge their contrast to a subjectively acceptable level. For a given defective video, we introduce an adaptive spatio-temporal connective (ASTC) filter, which adapts from temporal to spatial filters based on noise level and local motion characteristics to remove mixture of Gaussian and impulse noises. Both the temporal and the spatial filters are noniterative trilateral filters, formed by introducing a novel local image statistic—neighborhood connective value (NCV) into the traditional bilateral filter. NCV represents the connective strength of a pixel to all its neighboring pixels and is a good measure for differentiating between impulse noises and fine features. After noise removal, we adopt pyramid segmentation algorithm [13] to divide a frame into several regions. Based on the areas and standard deviations of these regions, we produce a novel adaptive piecewise mapping function (APMF) to automatically enhance the video contrast. To show effectiveness of our NCV statistic, we conducted a simulation experiment by adding impulse noises into three representative pictures and reported superior noise detection performance compared with other noise filters. In addition, we tested our system on several real defective videos adding mixed noises. These videos cover diverse kinds of defectiveness: underexposure, overexposure, mixture of them, and so forth. Our outputs are much more visually pleasing than those of other state-of-art approaches.

In addition, Section 2 reviews previous work related to video enhancement; the system framework is represented in Section 3; Section 7 gives the experimental results, followed by conclusions in Section 8.

There have been much previous work on image and video noise filter and contrast enhancement. We will make a brief review on this section and describe their essential differences with our work.

The input to our video enhancement system is a defective video mixed with Gaussian and impulse noises and having a visually undesirable contrast. We assume that the input video V is generated by adding the Gaussian noise G and impulse noises I to a latent video L. Thus, the input video can be represented by . Given the input defective video, the task of video enhancement system is to automatically generate an output video , which has visually desirable contrast and less noise. The system can be represented by a noise removal process f ₂ and a contrast enhancement process f ₁ as

(1)

Figure 1 illustrates the system framework of our video enhancement system. Like [21], we first extract the luminance and the chrominance of each frame, and then process the frame in luminance channel. To filter mixed noises in a given video, firstly a new local statistic—neighborhood connective value (NCV) is introduced to identify impulse noises, and then we incorporate it into the bilateral filter to form the spatial connective trilateral (SCT) filter and the temporal connective trilateral (TCT) filter. Then, we build an adaptive spatio-temporal connective (ASTC) filter adapting from TCT to SCT based on noise level and local motion characteristics. In order to deal with the presence of camera and object motion, our ASTC filter utilizes dense optical flows for motion compensation. Since typical optical flow techniques depend on robust gradient estimates and would fail on noisy low-contrast frames, we pre-enhance each frame by SCT filter and the adaptive piecewise mapping function (APMF).

In contrast enhancement procedure, we firstly separate a frame into large scale features and details using rank-ordered absolute difference (ROAD) bilateral filter [2], which preserves more fine features than other traditional filters do [26]. Then, we enhance the large scale features with APMF to achieve the desired contrast, while mapping the details using a less curved function adjusted by the local intensity standard deviation. This two pipeline method can avoid ringing artifacts even around sharp transition regions. Unlike traditional enhancement methods based on histogram statistics, we produce our adaptive piecewise mapping function (APMF) based on frame segmentation results, which provide more 2-D spatial information. Finally, the mapped large scale features, mapped details, and chrominance are combined to generate the final enhanced video. We next describe the NCV statistic, the ASTC noise filter, as well as the contrast enhancement procedure.

As shown in Figure 2(a), the pixels in the tiny lights are neither similar to most neighboring pixels [2] nor having small gradients in at least 4 directions [27], and thus will be misclassified as noises by [2, 27]. Comparing signal pixels in Figure 2(a) and noise pixels in Figure 2(b), we adopt the robust assumption that impulse noise pixels are always closely connected with fewer neighboring pixels than signal pixels [8]. Based on this assumption, we introduce a novel local statistic for impulse noise detection—neighborhood connective value (NCV), which measures the "connective strength" of a pixel to all the other pixels in its neighborhood window. In order to introduce NCV clearly, we need to make some important definitions first. In the following parts, let p _xy denotes the pixel with coordinates in a frame, and v _xy denotes its intensity.

Definition 1.

For two neighboring pixels p _xy and p _ij satisfying , their connective value (CV) is defined as

(2)

where equals 1 when , and equals 0.5 when is a parameter to penalize highly different intensities and is fixed to 30 in our experiments. The CV of two neighboring pixels assumes values in (0, 1]; the more similar their intensities are, the larger their CV is. CV measures the number of pixels that two neighboring pixels contribute to each other's "connective strength." It is perceptional rational that diagonal neighboring pixels are less closely connected than the neighboring pixels which share one identical edge, so one multiplies a factor (i.e., ) of different values to discriminate the two types of connection relationship.

Definition 2.

A path P from pixel p _xy to pixel p _ij is a sequence of the pixels , where and are neighboring pixels . The path connective value (PCV) is the product of CVs of all neighboring pairs along the path P

(3)

PCV describes the smoothness of a path; the more similar the intensities of pixels in the path are, the larger the path's PCV is. PCV achieves the maximum 1 when all pixels in the path have identical intensity; thus, . It should be noticed that there are several paths between two pixels. For example, in Figure 3, the path from p ₁₂ to p ₃₃ can be or , which have PCVs of 0.0460 and 0.2497, respectively.

Although PCV well describes the smoothness of a path, it fails to give a measure for the smoothness between one pixel in the neighborhood window and the central pixel. Thus, we introduce the following definition.

Definition 3.

The local connective value (LCV) of a central pixel p _xy with the pixel p _ij in its neighborhood window is the largest PCV of all the paths from p _xy to p _ij

(4)

In the above definitions, the neighboring pixels are pixels in a window, denoted by , with p _xy as the center. In our experiments, k is fixed to 2. LCV of one specific pixel equals the PCV of the smoothest path from it to the central pixel and reflects the geometric closeness and photometric similarity of it with the central one. Apparently, .

Definition 4.

The neighborhood connective value (NCV) of a pixel p _xy is the sum of LCVs of all its neighboring pixels

(5)

NCV provides a measure of the "connective strength" of a central pixel to all its neighboring pixels. For a neighborhood window, NCV will decrease to about 1 when the intensity of the central pixel far deviates from those of all neighboring pixels and will reach its maximum 25, when all the pixels in the neighborhood window have identical intensity, so .

To get NCV, LCV must be calculated first. In order to compute LCV more easily, one needs to make some mathematical transform first:

(6)

Let , and one has

(7)

Since , then one has . Thus, one can make a graph, taking the central pixel and all its neighboring pixels as vertices and taking DIS as the cost of edge between two pixels. Therefore, the calculation of LCV can be converted to the single-source shortest path problem and can be solved by Dijkstra's algorithm [28].

To test the effectiveness of NCV for impulse noise detection, one conducted a simulation experiments on three representative pictures: "Lena," "Bridge," and "Neon Light" as shown in Figure 4. "Lena" has few sharp transitions, "Bridge" has many edges, and "Neon Light" has lots of impulse-like fine features, that is, small high gradient regions. The diverse characteristics of these pictures assure the effectiveness of our experiments. Figures 5(a), 5(b), and 5(c) display quantitative results from the "Lena," "Bridge," and "Neon Light" images, respectively. The lower dashed lines represent the mean NCV for salt-and-pepper noise pixels—which is a discrete impulse noise model in which the noisy pixels take only the values 0 and 255—as a function of the amount of noise added, and the upper dashed line represents the mean NCV for signal pixels. The signal pixels consistently have higher mean NCVs than the impulse pixels, of which NCVs remain almost constant even with very high noise level. In contrast, the famous ROAD statistic cannot well differentiate between impulse and signal pixels in the "Neon Light" image as shown in Figure 5(d), because it assumes the signal pixels have at least half similar pixels in neighborhood window, which is coincident with the smooth regions but corrupts for fine features.

In order to enhance the NCV's ability of noise detection, we map NCV to a new value domain and introduce the inverted NCV as

(8)

Thus, INCVs of impulse pixels will fall into large value ranges, whereas those of signal pixels will cluster near zero. Obviously, .

Video is a compound of image sequences, including both spatial and temporal information. Accordingly, our ASTC video noise filter adapts from temporal to spatial noise filter. We will detail the spatial filter, the temporal filter, and the adaptive fusion strategy in this section.

5.1. The Spatial Connective Trilateral Filter

As mentioned in Section 4, NCV is a good statistic for impulse noise detection, whereas the bilateral filter [2] well suppresses Gaussian noise. Thus, we incorporate NCV into the bilateral filter to form a trilateral filter in order to remove mixed noise.

For a pixel p _xy , its new intensity after bilateral filtering is computed as

(9)

(10)

where and represent spatial and radiometric weights, respectively [2]. In our experiments, and are fixed to 2 and 30, respectively. The formula is based on the assumption that pixels locating nearer and having more similar intensities should have larger weights.

As to images with noises, intuitively, the signal pixels should have larger weights than the noise pixels. Thus, similar to the above, we introduce a third weighting function to measure the probability of a pixel being a signal pixel:

(11)

Where is a parameter to penalize large INCVs and is fixed to 0.3 in our experiments. Thus, we can integrate into (10) to form a better weighting function. Yet, direct integration will fail to process impulse noise pixels because neighboring signal pixels will have lower than other impulse pixels of similar intensity. As a result, the impulse pixels remain impulse pixels. To solve this problem, Garnett et al. [2] brought forward a switch function J to determine the weight of the radiometric component in the presence of impulse noise. Similarly, our switch is defined as

(12)

The switch J tends to reach its maximum 1, when p _xy or p _ij has large INCV, that is, with high probability of being a noise pixel; J tends to reach its minimum 0, when both p _xy and p _ij have small INCVs, that is, with high probability of being signal pixels. Thus, we introduce the switch J into (10) to control the weights of and as

(13)

According to the new weighting function, for impulse noise pixels, is almost "shut off" by the switch J, while and work to remove the large outliers; for other pixels, is almost "shut off" by the switch J, and only and work to smooth small amplitude noise without blurring edges. Consequently, we build the spatial trilateral connective (SCT) filter by merging (9) and (13).

Figure 6 shows the outputs of ROAD and SCT filters for the "Neon Light" image corrupted by mixed noise. ROAD filter is based on a rank-order statistic for impulse detector and the bilateral filter. It can well smooth the mixed noise with PSNR = 23.35 but blur lots of fine features such as the tiny lights in Figure 6(b). In contrast, our SCT filter preserves more fine features and produces more visually pleasing output with PSNR = 24.13, as shown in Figure 6(c).

5.2. Trilateral Filtering in Time

As to videos, temporal filtering is more important than spatial filtering [10], but irregular camera and object motions often degrade the performance. Thus, robust motion compensation is quite necessary. Optical flow is a classical approach for this problem; however, it depends on robust gradient estimation and will fail for noisy, underexposed, or overexposed images. Therefore, we pre-enhance the frames with SCT filter and our adaptive piecewise mapping function, which will be detailed in Section 6. Then, we adopt the cvCalcOpticalFlowLK() function of the intel open source computer vision library (Opencv) to compute dense optical flows for robust motion estimation. Too small and too large motions are deleted; also, half-wave rectification and Gaussian smoothing are applied to eliminate noises in optical flow field [29].

After motion compensation, we adopt the similar approach to SCT filter in temporal direction. In temporal connective trilateral (TCT) filter, we define the neighborhood window of a pixel p _xyt as , which is a ( )-length window in temporal direction with p _xyt as the middle. In our experiments, m is fixed to 10. Noticing that the pixels in the window may have different horizontal and vertical coordinates in frames, but they are on the same tracking path generated by the optical flow. Thus, the TCT filter is computed as

(14)

where and and J are defined the same as (11) and (12), respectively.

The TCT filter can well differentiate impulse noise pixels from motional pixels and smooth the former while leaving the later almost untouched. For impulse noise pixels, the switch function J in TCT filter will "shut off" the radiometric component and the spatial weight is used to smooth them; for motional pixels, J will "shut off" the impulsive component and TCT filter reverts to bilateral filter, which takes the motional pixels as "temporal edges" and leaves them unchanged.

5.3. Implementing ASTC

Although TCT filter is based on robust motion estimation, there are often not enough similar pixels in temporal direction for smoothing in presence of complex motions. As a result, the TCT filter fails to achieve desirable smoothing results and have to convert to spatial direction. Thus, a threshold is necessary to determine whether a sufficient number of temporal similar pixels are gathered; this threshold then can be used as a switch between temporal and spatial filters (in [21]), or as a parameter adjusting importance of the two filters (in our ASTC). If the threshold is too high, then for severely noisy videos, there are always not enough valuable temporal pixels, and temporal filter becomes useless; if the threshold is too low, then no matter how noisy a video is, the output will be always based on unreliable temporal pixels. Accordingly, we introduce an adaptive threshold like [21], but further considering local noise levels:

(15)

In the above formula, presents the local noise level and is computed in a spatial neighborhood window. reaches its maximum 1 in good frames and decreases with the increase of noise level. is the gain factor of current pixel and equals the tone mapping scales in our adaptive piecewise mapping function, which will be detailed in Section 6. Thus, the more mapping scale is and less noises exist, the larger becomes; the less mapping scale is and more noises exist, the smaller becomes. Such characteristics assure the threshold working well for different kinds of videos.

Since the temporal filter outperforms the spatial filter when gathering enough temporal information, we propose the following criteria for the fusion of temporal filter and spatial filter.

1. (1)

   If a sufficient number of temporal pixels are gathered, only temporal filter is used.

    
2. (2)

   On the other hand, even if temporal pixels are insufficient, the temporal filter should still more dominant over the spatial one in the fused spatio-temporal filter.

    

Based on these two criteria, we propose our adaptive spatio-temporal connective (ASTC) filter, which adaptively fuses the spatial connective trilateral filter and temporal connective trilateral filter as

(16)

where

(17)

which represents the sum of pixel weights in temporal direction. If (i.e., sufficient temporal pixels), , then ASTC filter regresses to temporal connective trilateral filter; if (i.e., insufficient temporal pixels), , ASTC filter will use the temporal connective trilateral filter to gather pixels in temporal direction first, and then use the spatial connective trilateral filter to gather the remaining number of pixels in spatial direction.

We have described the process of filtering mixture of Gaussian and impulse noises from defective videos. However, contrast enhancement is another key issue. In this section, we will show how to build the tone mapping function as well as how to automatically adjust important parameters and smooth the function in time.

6.1. Generating AMPF

As the target of our video enhancement system is to deal with diverse videos, our tone mapping function needs to work well for videos corrupted by underexposure, overexposure, or mixture of them. Thus, a piecewise mapping function is needed to treat these two kinds of ill-exposed pixels differently. As shown in Figure 7, we divide our mapping function into low and high segments according to a threshold , and each segment adapts its curvature individually. In order to get a suitable , we introduce two threshold values, Dark and Bright; [0, Dark] denotes the dark range, and [Bright, 1] denotes the bright range. According to human's perception, we set Dark and Bright to 0.1 and 0.9, respectively. Perceptively, if there are more pixels falling into dark range than those into bright range, we should use low segment more and assign a larger value. On the other hand, if there are much more pixels falling in bright range, we should use high segment more and assign a smaller value. A simple approach to determine is to use pixel numbers in Dark and Bright areas. Yet, owing to our APMF is calculated before the ASTC filter, there are still somewhat noises, and pixel numbers are not quite reasonable. Thus, we use the pyramid segmentation algorithm [13] to segment a frame into several connected regions and use the region area information to determine . Let , and denote the area, the average intensity, and the standard deviation of intensities of the i th region, respectively. Then, we compute by

(18)

If is larger than Bright, then it is assigned to 1, and the low-segment curve will occupy the whole dynamic range; if is lower than Dark, then it is assigned to 0, and the high-segment curve will occupy the whole dynamic range. If there are no regions with average intensities falling into either dark or bright range, then is assigned to the default value 0.5.

With division of intensity range, the tone mapping function can be designed separately for low and high segments. Considering human perception responses, Bennett and McMillan [21] proposed a logarithmic mapping function, which well deals with underexposed videos. We incorporate their function to our adaptive piecewise mapping function (APMF) in underexposed areas but extended the function to also deal with overexposed areas as follows:

(19)

where and are parameters controlling the curvatures of low and high segments, respectively. and are gain factors of intensities Dark and Bright, respectively, which is defined the same as in (15), that is, the proportion between the new intensity and the original one. and are precomputed before getting the mapping function and control the selection of curves between the red and the green in Figure 7. This mapping function avoids sharp slope near the origin, and thus well preserves details [21].

6.2. Automatic Parameters Selection

Although we designed the APMF as (19) to deal with different situations, how to choose appropriate parameters in the function determines the tone mapping performance. Thus, we will detail the process of choosing these important parameters— , and .

When certain dynamic range is enlarged, there must be some other ranges being compressed. As to an intensity range , if more segmented regions fall into it, then there is probably more information in this range, and thus the contrast should be enlarged, that is enlarging the intensity range. On the other hand, if the standard deviation of regions in this range is quite large, then it is probably that the contrast is already enough and needs not to be enlarged anymore [30]. According to the above, we define the enlarged range R of as

(20)

where N is the normalization operator (divided by the maximum), and I is the maximum range which can be stretched to. In other words, denotes the maximum enlarging range, and the exponential factor controls the enlarging scale. It should be noticed that the segmented regions with too small standard deviation should be disregarded in (20) because they probably correspond to the backgrounds or monochromic boards in the image and should not be enhanced anymore.

We take the low segment curve in Figure 7 as an example. If [0, Dark] is enlarged, the red curve should be adopted, and Dark is extended to . The maximum of is , and thus can be represented as R (0, Dark, ). Similarly, if [Dark, ] is enlarged, the green curve should be adopted, and Dark is compressed to , in which l ₂ is represented as . Therefore, considering both parts, we make the new mapping intensity of Dark as . Then is , and can be computed by solving the following equation:

(21)

and can be got similarly. Thus, all the parameters in (19) are determined.

As mentioned in Section 2, in order to better deal with details as well as avoiding ringing artifacts, we first separate an image into large scale parts and details using ROAD bilateral filter owing to its ability of well preserving fine features [26], and then enhance the large scale parts with function , while enhancing details with a less curved function . and correspond to the intensity standard deviations of all regions falling into and , respectively. The larger the standard deviation is, the more linear the mapping function for the details is.

APMF can also avoid introducing washed-out artifacts, that is, over enhancing images with homochromous backgrounds. Figure 8(a) shows an image of moon with black background. The histogram equalization result exhibits a washed-out appearance shown in Figure 8(b), for the reason that the background corresponds to the largest component in histogram and causes the whole picture enhanced too much [12]. Figure 8(c) shows the result of the most popular image processing software, Photoshop, using its "Auto Contrast" function [31]. The disappointing appearance comes from its disregarding the first 0.5% of the range of white and black pixels, which leads to loss of information in the clipped ranges. Figure 8(d) shows the APMF result, and we can see that the craters in the central of image are quite clear.

6.3. Temporal Filtering of AMPF

APMF is formed based on the statistical information of each frame separately, and differences contained in the successive frames may result in disturbing flicker. Small difference means that the scene of video is very smooth and the flicker can be reduced by smoothing the mapping functions. Large difference probably means that a shot cut occurring and the current mapping function should be replaced by a new one. Since APMF is determined by three values— , and , we define the function difference as

(22)

where is the difference operator. If Diff of successive frames is lower than a threshold, then we smooth , and in the APMF of current frame by averaging corresponding values in neighboring ( ) frames. Otherwise, we just adopt the new APMF. In our experiments, m is fixed to 5 and the threshold is 30.

To demonstrate the effectiveness of the proposed video enhancement framework, we have applied it to a broad variety of low-quality videos, including corrupted by mixed Gaussian and impulse noise, underexposed and overexposed video sequences. Although it is difficult to obtain the ground truth comparison for video enhancement, it can be clearly seen from the processed results that our framework is superior to the other existing methods.

First, we compare performances of our video enhancement system with ASTA system. Since ASTA can only work for underexposed videos, we only do the comparison on such videos. In addition, we also make comparisons with other two most common 3-dimensional median filters—P3D [32] filter and AML3D [33] filters followed by histogram equalization and our APMF. The results are shown in Figures 9, 10, and 11, which are experiments on underexposed video, overexposed video, and video with under- and over-exposed regions. Since underexposed regions are assumed owning zero-mean Gaussian noise [21], we only add uniformly distributed impulse noise to such videos as shown in Figures 9(a), 11(a), but add mixed noise to over-exposed video as shown in Figure 10(b).

From all picture (b), (c) of Figures 9, 10, and 11, all of which are enhanced by the popular contrast enhancement method-histogram equalization, we can see that no matter whether the noises are filtered in advance (all Figure (c)) or not (all Figure (b)), the output videos are always unacceptable, since the noises are over-enhanced in the equalization process. While our APMF considers the intensity standard deviations and treat large scale parts and details differently. From Figures 10(c), 10(d), 11(c), and 11(d), we can see that our APMF produces much better outputs than histogram equalization after the same filtering process. Our APMF great enhances the video as well as suppressing mixed noises. In addition, our APMF produces desirable outputs in all underexposed, overexposed, and mixed ill-exposed videos, owing to its ability of adaptively adjusting the mapping functions according to different videos.

As to noise filtering, our ASTC filter also outperforms other approaches. Although the ASTA system work well on videos with Gaussian noises [21], it fails to deal with videos with mixed noises as shown in Figure 9(d). We can see great impulse noise pixels allover the image. This is because ASTA is formed by combining the spatial and temporal bilateral filters, which take the impulse noise pixels as "temporal edges" and leave them untouched. In addition, AML3D filter and P3D filter, which are two kinds of improved spatio-temporal median filters, produce grainy results as shown in the bright wall regions in Figure 10(d) as well as the dark regions in Figure 11(d). In contrast, our system produces more pleasing outputs as shown in Figure 10(e) and well preserves details that are hardly visible in the original videos such as the car in Figure 9(e) and the telephone in Figure 11(e). The reason is that our noise filter is based on the combination of a good impulse detector and the classical bilateral filter; the former well deals with large outliers, and the latter effectively smoothes small amplitude noises. In general, the results indicate the robustness and effectiveness of our video enhancement system in different kinds of videos with mixed noises.

In this paper, we have presented a universal video enhancement system, which is able to greatly suppress the most two common noises—Gaussian and impulse noises as well as significantly enhance video contrast. We introduce a novel local image statistic—neighborhood connective value (NCV) to improve impulse noise detection performance to a great extent. Then, we incorporate it into the bilateral filter framework to form an adaptive spatio-temporal connective (ASTC) filter to reduce mixed noises. ASTC filter adapts from a temporal filter to a spatial one based on noise level and local motion characteristics, and thus assure its robustness for different videos. Furthermore, we build an adaptive piecewise mapping function (APMF) to automatically enhance video contrast using statistical information of frame segmentation results, which provide more 2-D spatial information than the histogram statistics. We conducted a simulation experiment on three representative images, and an extensive experiment on several videos, which are underexposed, overexposed, or both of them. Both the objective and subjective evaluations indicated the effectiveness of our system.

Limitations remain in our system, however. First, our system assumes that impulse noise pixels are always closely connected with fewer neighboring pixels than signal pixels, so it will fail to remove large blotches (i.e., distorted region larger than four pixels) for film restoration. Secondly, our implementation is very slow since it includes multiple nonlinear filtering steps and computation of NCVs. The current processing of one frame takes about one minute. Extending our approach to detect large blotches and improving its performance are our future work. Furthermore, we will pay attention to enhance video regions differently according to human's attention model.